CN116259012B - Monitoring system and method for embedded supercharged diesel tank - Google Patents
Monitoring system and method for embedded supercharged diesel tank Download PDFInfo
- Publication number
- CN116259012B CN116259012B CN202310547149.0A CN202310547149A CN116259012B CN 116259012 B CN116259012 B CN 116259012B CN 202310547149 A CN202310547149 A CN 202310547149A CN 116259012 B CN116259012 B CN 116259012B
- Authority
- CN
- China
- Prior art keywords
- tank
- feature
- diesel
- context
- diesel oil
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 153
- 238000000034 method Methods 0.000 title claims abstract description 47
- 239000002283 diesel fuel Substances 0.000 claims abstract description 113
- 238000012545 processing Methods 0.000 claims abstract description 41
- 239000013598 vector Substances 0.000 claims description 127
- 239000011159 matrix material Substances 0.000 claims description 117
- 238000013527 convolutional neural network Methods 0.000 claims description 39
- 238000007781 pre-processing Methods 0.000 claims description 31
- 238000000605 extraction Methods 0.000 claims description 14
- 238000005457 optimization Methods 0.000 claims description 10
- 230000004913 activation Effects 0.000 claims description 9
- 238000011176 pooling Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 7
- 230000007246 mechanism Effects 0.000 claims description 7
- 230000005540 biological transmission Effects 0.000 claims description 6
- 238000000638 solvent extraction Methods 0.000 claims 1
- 238000004458 analytical method Methods 0.000 abstract description 10
- 238000013135 deep learning Methods 0.000 abstract description 6
- 238000013473 artificial intelligence Methods 0.000 abstract description 4
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 20
- 238000009826 distribution Methods 0.000 description 16
- 238000004590 computer program Methods 0.000 description 11
- 230000006870 function Effects 0.000 description 11
- 238000003860 storage Methods 0.000 description 10
- 230000008569 process Effects 0.000 description 8
- 230000008901 benefit Effects 0.000 description 5
- 239000000446 fuel Substances 0.000 description 4
- 238000003062 neural network model Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 208000027697 autoimmune lymphoproliferative syndrome due to CTLA4 haploinsuffiency Diseases 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000001149 cognitive effect Effects 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 238000010191 image analysis Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000005012 migration Effects 0.000 description 2
- 238000013508 migration Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013139 quantization Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000005553 drilling Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Examining Or Testing Airtightness (AREA)
Abstract
A monitoring system and method of embedded supercharged diesel oil tank, it obtains the monitoring image of the diesel mother tank of the embedded supercharged diesel oil tank gathered by the camera; and (3) performing image processing and analysis on the outer surface image of the diesel oil master tank by adopting an artificial intelligence technology based on deep learning, and generating an early warning prompt whether leakage occurs or not based on the image processing and the analysis. Thus, the leakage monitoring can be intelligently performed on the diesel oil master tank so as to ensure the running stability of the embedded supercharged diesel oil tank.
Description
Technical Field
The application relates to the technical field of intelligent monitoring, and in particular relates to a monitoring system and a monitoring method of an embedded supercharged diesel tank.
Background
Chai Youguan is used as fuel storage equipment of field operation power equipment such as well drilling, and each operation unit is provided, and because the on-site diesel engine is provided with a longer distance from a diesel tank, the fuel pipeline is longer, and when the liquid level in the tank is lower, the fuel supply pressure of the engine is smaller, so that the poor fuel supply of the engine is caused, and the normal operation of the engine is influenced.
In order to solve the problems, the diesel oil tanks are generally made into two different sizes and are overlapped to be communicated with a pipeline for use, the lower tank is used as an oil storage tank, and the upper tank is used as an oil supply tank, so that the use of field equipment is satisfied due to the fact that the oil supply tank and oil utilization equipment always have a fall.
The disadvantages are as follows: 1. when the upper tank is moved every time, the lifting operation is required to be installed and detached, the safety risk exists, 2, the upper tank and the lower tank are externally arranged in a communicated pipeline, the leakage condition exists, and the environment protection requirement is not met. 3. When the vehicle is moved, the pipeline communicated with the upper tank and the lower tank needs to be disassembled, diesel oil in the pipeline leaks, and meanwhile, the diesel oil is polluted. 4. The transportation cost is increased, and two vehicles are required for transporting the upper tank and the lower tank.
Thus, an optimized pressurized diesel tank and monitoring scheme therefor is desired.
Disclosure of Invention
The present application has been made in order to solve the above technical problems. The embodiment of the application provides a monitoring system and a monitoring method of an embedded type supercharged diesel tank, wherein the monitoring system and the monitoring method acquire monitoring images of a diesel master tank of the embedded type supercharged diesel tank, which are acquired by a camera; and (3) performing image processing and analysis on the outer surface image of the diesel oil master tank by adopting an artificial intelligence technology based on deep learning, and generating an early warning prompt whether leakage occurs or not based on the image processing and the analysis. Thus, the leakage monitoring can be intelligently performed on the diesel oil master tank so as to ensure the running stability of the embedded supercharged diesel oil tank.
In a first aspect, a monitoring system for an embedded pressurized diesel tank is provided, comprising:
The image acquisition module is used for acquiring a monitoring image of the diesel master tank of the embedded type supercharged diesel tank acquired by the camera;
the image preprocessing module is used for preprocessing the image of the monitoring image of the diesel oil master tank to obtain a preprocessed monitoring image;
the feature extraction module is used for enabling the preprocessed monitoring image to pass through a convolutional neural network model serving as a feature extractor to obtain a diesel oil mother tank surface feature matrix;
the characteristic matrix dividing module is used for dividing the characteristic matrix of the surface of the diesel oil mother tank to obtain a plurality of local characteristic matrixes;
the context coding module is used for expanding the local feature matrixes into a plurality of local expansion feature vectors and then obtaining a global feature matrix on the surface of the diesel oil mother tank through a context coder based on a converter; and
the early warning prompt generation module is used for enabling the global feature matrix on the surface of the diesel oil mother tank to pass through a classifier to obtain a classification result, and the classification result is used for indicating whether leakage early warning prompts are generated or not.
In the above monitoring system of an embedded supercharged diesel tank, the image preprocessing module is configured to: and carrying out image preprocessing on the monitoring image of the diesel oil master tank by adopting a filter and histogram equalization to obtain the preprocessed monitoring image.
In the above monitoring system of an embedded supercharged diesel tank, the feature extraction module is configured to: and respectively carrying out convolution processing, pooling processing along a channel dimension and nonlinear activation processing on input data in forward transmission of layers by using each layer of the convolutional neural network model serving as a feature extractor to output the final layer of the convolutional neural network model serving as the feature extractor as the diesel tank surface feature matrix, wherein the input of the first layer of the convolutional neural network model serving as the feature extractor is the preprocessed monitoring image.
In the above monitoring system for an embedded supercharged diesel tank, the feature matrix dividing module is configured to: and uniformly dividing the surface feature matrix of the diesel oil mother tank to obtain a plurality of local feature matrices.
In the above-mentioned monitoring system of embedded supercharged diesel tank, the context coding module includes: a self-attention unit for performing context semantic coding based on a self-attention mechanism on the plurality of local expansion feature vectors by using the context encoder based on the converter to obtain a plurality of context local expansion feature vectors; a gaussian regression factor calculation unit for calculating a gaussian regression uncertainty factor for each of the plurality of contextual locally-expanded feature vectors; the weighting unit is used for respectively weighting each context local expansion feature vector by taking the Gaussian regression uncertainty factor of each context local expansion feature vector as a weight so as to obtain a plurality of weighted context local expansion feature vectors; and the two-dimensional arrangement unit is used for carrying out two-dimensional arrangement on the plurality of weighted context local expansion feature vectors to obtain the diesel oil master tank surface global feature matrix.
In the above monitoring system for an embedded supercharged diesel tank, the gaussian regression factor calculation unit is configured to: calculating a gaussian regression uncertainty factor for each of the plurality of contextual locally-expanded feature vectors with an optimization formula; wherein, the optimization formula is:
，
wherein,,is the length of the feature vector, +.>And->The mean and variance of feature sets at each location in the plurality of contextual locally expanded feature vectors, respectively, wherein +.>Is the +.f of the local expansion feature vector of the plurality of contexts>Characteristic value of individual position, and->As a logarithmic function with base 2 +.>Is the gaussian regression uncertainty factor.
In the above-mentioned monitoring system of embedded pressure boost diesel tank, early warning suggestion generation module includes: the matrix unfolding unit is used for unfolding the global feature matrix on the surface of the diesel oil mother tank into classification feature vectors according to row vectors or column vectors; the full-connection coding unit is used for carrying out full-connection coding on the classification characteristic vectors by using a plurality of full-connection layers of the classifier so as to obtain coded classification characteristic vectors; and the classification unit is used for passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
In a second aspect, a method for monitoring an embedded supercharged diesel tank is provided, which comprises the following steps:
acquiring a monitoring image of a diesel master tank of an embedded pressurized diesel tank acquired by a camera;
performing image preprocessing on the monitoring image of the diesel oil master tank to obtain a preprocessed monitoring image;
the preprocessed monitoring image passes through a convolutional neural network model serving as a feature extractor to obtain a diesel oil mother tank surface feature matrix;
performing feature matrix division on the surface feature matrix of the diesel oil mother tank to obtain a plurality of local feature matrices;
expanding the local feature matrixes into a plurality of local expansion feature vectors, and then passing through a context encoder based on a converter to obtain a global feature matrix on the surface of the diesel tank; and
and the global feature matrix on the surface of the diesel oil mother tank passes through a classifier to obtain a classification result, wherein the classification result is used for indicating whether leakage early warning prompt is generated or not.
In the above monitoring method of an embedded supercharged diesel tank, performing image preprocessing on a monitoring image of the diesel tank to obtain a preprocessed monitoring image, including: and carrying out image preprocessing on the monitoring image of the diesel oil master tank by adopting a filter and histogram equalization to obtain the preprocessed monitoring image.
In the above monitoring method of an embedded supercharged diesel tank, passing the preprocessed monitoring image through a convolutional neural network model as a feature extractor to obtain a diesel mother tank surface feature matrix, including: and respectively carrying out convolution processing, pooling processing along a channel dimension and nonlinear activation processing on input data in forward transmission of layers by using each layer of the convolutional neural network model serving as a feature extractor to output the final layer of the convolutional neural network model serving as the feature extractor as the diesel tank surface feature matrix, wherein the input of the first layer of the convolutional neural network model serving as the feature extractor is the preprocessed monitoring image.
Compared with the prior art, the monitoring system and the method for the embedded type supercharged diesel tank, which are provided by the application, acquire the monitoring image of the diesel master tank of the embedded type supercharged diesel tank, which is acquired by the camera; and (3) performing image processing and analysis on the outer surface image of the diesel oil master tank by adopting an artificial intelligence technology based on deep learning, and generating an early warning prompt whether leakage occurs or not based on the image processing and the analysis. Thus, the leakage monitoring can be intelligently performed on the diesel oil master tank so as to ensure the running stability of the embedded supercharged diesel oil tank.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments or the description of the prior art will be briefly introduced below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic diagram of an embedded pressurized diesel tank according to an embodiment of the present application.
Fig. 2 is an application scenario diagram of a monitoring system of an embedded supercharged diesel tank according to an embodiment of the present application.
Fig. 3 is a block diagram of a monitoring system for an embedded pressurized diesel tank according to an embodiment of the present application.
Fig. 4 is a block diagram of the context encoding module in the monitoring system of the embedded pressurized diesel tank according to an embodiment of the present application.
Fig. 5 is a block diagram of the early warning prompt generation module in the monitoring system of the embedded supercharged diesel tank according to the embodiment of the application.
Fig. 6 is a flowchart of a method of monitoring an embedded pressurized diesel tank according to an embodiment of the present application.
Fig. 7 is a schematic diagram of a system architecture of a monitoring method of an embedded supercharged diesel tank according to an embodiment of the present application.
1, a diesel oil mother tank; 2. chai Youzi cans; 3. a pump house; 4. a diesel pump, 5, a diesel suction pipe; 6. a diesel oil discharge pipe; 7. and a diesel oil taking port.
Detailed Description
The following description of the technical solutions in the embodiments of the present application will be made with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
Unless defined otherwise, all technical and scientific terms used in the examples of this application have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to limit the scope of the present application.
In the description of the embodiments of the present application, unless otherwise indicated and defined, the term "connected" should be construed broadly, and for example, may be an electrical connection, may be a communication between two elements, may be a direct connection, or may be an indirect connection via an intermediary, and it will be understood by those skilled in the art that the specific meaning of the term may be understood according to the specific circumstances.
It should be noted that, the term "first\second\third" in the embodiments of the present application is merely to distinguish similar objects, and does not represent a specific order for the objects, it is to be understood that "first\second\third" may interchange a specific order or sequence where allowed. It is to be understood that the "first\second\third" distinguishing objects may be interchanged where appropriate such that the embodiments of the present application described herein may be implemented in sequences other than those illustrated or described herein.
In view of the above technical problems, the technical concept of the present application is to provide an embedded booster Chai Youguan, as shown in fig. 1, which includes a diesel main tank 1, a diesel pump 4, a diesel sub-tank 2, a diesel oil intake port 7, a diesel oil intake pipe 5, a pump house 3 and a diesel oil discharge pipe 6, wherein diesel oil is first loaded into the diesel main tank 1, and diesel oil in the diesel main tank 1 is transferred into the diesel sub-tank 2 by the diesel pump 4. The oil using equipment is used for taking oil from the diesel sub-tank 2 through the diesel oil taking port 7 so as to supply the diesel oil with pressure to the oil using equipment. When the use of diesel oil in the diesel oil sub-tank 2 is reduced to a certain height, the diesel oil pump 4 is started to supplement the diesel oil in the diesel oil main tank 1 into the diesel oil sub-tank 2, and the above process is repeated to supply the pressurized diesel oil to the oil using equipment.
The beneficial effects of the production are as follows: 1. the oil storage tank and the oil supply tank are integrated together, so that the oil storage tank and the oil supply tank are not required to be installed in field use, and the installation time is shortened. 2. Because the hoisting device is used on site without installation, the safety risk existing in hoisting operation is avoided. 3. Compared with the prior art, the external pipeline is reduced, the dripping phenomenon is avoided, and the environment is protected. 4. Compared with the prior art, the equipment has the advantages of smaller volume and reduced pulling and transporting cost.
In the operation process of the embedded supercharged diesel oil tank system, the embedded supercharged diesel oil tank is required to be subjected to leakage monitoring so as to ensure that the embedded supercharged diesel oil tank is stable in operation. That is, in the technical scheme of the application, a monitoring system of an embedded supercharged diesel tank is also provided, which is used for monitoring leakage of the diesel tank so as to ensure the running stability of the embedded supercharged diesel tank.
Aiming at the technical requirements, the technical conception of the application is as follows: the external surface image of the diesel oil master tank is acquired through the camera, and then the external surface image of the diesel oil master tank is subjected to image processing and analysis, so that the leakage detection is realized. Here, it should be understood that if the diesel tank leaks, and oil stains may occur on the outer surface, so that the image of the outer surface of the diesel tank changes, it may be determined whether the diesel tank leaks based on the change of the image end.
Specifically, in the technical scheme of the application, firstly, a monitoring image of a diesel master tank of an embedded type supercharged diesel tank collected by a camera is obtained. It will be appreciated that in the embedded pressurized diesel tank system, the diesel tank is an integral part of the diesel storage in the system, and therefore monitoring the diesel tank is very important. Correspondingly, the monitoring image of the diesel oil master tank is acquired by the camera, so that the condition of the surface of the diesel oil master tank can be observed in real time.
And then, carrying out image preprocessing on the monitoring image of the diesel fuel mother tank to obtain a preprocessed monitoring image. In practical applications, the acquired image may contain various noises, such as uneven brightness, artifacts, shadows, etc., due to uncertainty of the camera environment and interference of other factors. These noises can affect the accuracy of subsequent computer vision and deep learning techniques, and therefore require pre-processing of the image to remove or reduce these disturbances. Accordingly, by preprocessing the monitored image, operations such as background noise elimination, edge enhancement, contrast enhancement and the like can be performed, so that the extracted features are more accurate and stable. For example, a filter may be employed to smooth the image to eliminate noise; histogram equalization may also be used to adjust the brightness and contrast of the image.
After image preprocessing, the preprocessed monitoring image is passed through a convolutional neural network model serving as a feature extractor to obtain a diesel tank surface feature matrix. Namely, the convolution neural network model with excellent performance in the field of image feature extraction is used for carrying out image local feature extraction based on convolution kernels on the preprocessed monitoring image so as to obtain the diesel oil master tank surface feature matrix.
Those of ordinary skill in the art will appreciate that the Convolutional Neural Network (CNN) is a deep learning model widely used in the field of computer vision that automatically learns feature representations from a large amount of data and efficiently processes two-dimensional data such as images. In the scheme, a pre-trained convolutional neural network model is used as a feature extractor, and convolution processing, pooling processing along the channel dimension and nonlinear activation processing are carried out on the extracted monitoring image to obtain the diesel oil master tank surface feature matrix.
In order to capture long-range dependency information among various local features, in the technical scheme of the application, firstly, feature matrix division is carried out on the diesel oil master tank surface feature matrix to obtain a plurality of local feature matrices, the local feature matrices are unfolded into a plurality of local unfolded feature vectors, and then the local feature matrices are unfolded through a context encoder based on a converter to obtain the diesel oil master tank surface global feature matrix.
Specifically, the feature matrix division is performed on the surface feature matrix of the diesel oil master tank, so that a large and high-dimensional feature matrix is divided into a plurality of smaller local feature matrices, and preamble support is provided for context coding among subsequent local feature matrices. Then, the local feature matrices are expanded into local expansion feature vectors, and then the local expansion feature vectors are obtained through a context encoder based on a converter, wherein the context encoder based on the converter utilizes a self-attention mechanism in the concept of the converter to capture global context association information of the local expansion feature vectors relative to the local expansion feature vectors so as to obtain the local expansion feature vectors. And after the plurality of context local expansion feature vectors are obtained, the plurality of context local expansion feature vectors are arranged into the diesel oil master tank surface global feature matrix.
And finally, the global feature matrix on the surface of the diesel oil master tank passes through a classifier to obtain a classification result, wherein the classification result is used for indicating whether leakage early warning prompt is generated or not. The classifier is used for determining a class probability label of the global feature matrix of the surface of the diesel tank, and the class probability label is used for indicating whether leakage early warning prompt is generated or not. Thus, an intelligent monitoring scheme of the diesel fuel mother tank is constructed based on machine vision and intelligent image analysis and processing calculation.
Specifically, in the technical scheme of the application, after the characteristic matrix of the surface of the diesel oil master tank is obtained through a convolution neural network model serving as a characteristic extractor and is divided by the characteristic matrix, gaussian distribution error uncertainty of characteristic distribution of each local characteristic matrix is caused, after context correlation coding of image semantic characteristics is carried out based on a context encoder of a converter, gaussian distribution error uncertainty of characteristic distribution of each characteristic vector in a plurality of context local expansion characteristic vectors is further carried out, and classification regression errors are caused in the global characteristic matrix of the surface of the diesel oil master tank obtained by two-dimensional arrangement of the plurality of context local expansion characteristic vectors, so that accuracy of classification results obtained by the classifier of the global characteristic matrix of the surface of the diesel oil master tank is affected.
Based on this, in the technical solution of the present application, a gaussian regression uncertainty factor of each of the plurality of context local expansion feature vectors is calculated, and is expressed as:
，
wherein,,is the length of the feature vector, +. >And->The mean and variance of feature sets at each location in the plurality of contextual locally expanded feature vectors, respectively, wherein +.>Is the +.f of the local expansion feature vector of the plurality of contexts>Characteristic value of individual position, and->As a logarithmic function with base 2 +.>Is the gaussian regression uncertainty factor.
Here, for the agnostic regression of the global feature matrix of the diesel tank surface, which may be caused by the distribution uncertainty information of the integrated feature set of each of the plurality of context local expansion feature vectors, the scalar measurement of the statistical characteristics of the feature set is performed by using the mean value and the variance as the statistical quantization parameters, so that the normal distribution cognitive mode represented by the feature errors is expanded to an unknown distribution regression mode, and the migration learning based on natural distribution transfer on the feature set scale is realized, so that the global feature matrix of the diesel tank surface is obtained by weighting each context local expansion feature vector by the gaussian regression uncertainty factors and then two-dimensionally arranging the weighted context local expansion feature vectors, and the uncertainty correction based on self calibration of each context local expansion feature vector when the global feature matrix of the diesel tank surface is formed, so that the classification regression errors existing in the global feature matrix of the diesel tank surface are corrected, and the accuracy of the classification result obtained by the classifier is improved.
Fig. 2 is an application scenario diagram of a monitoring system of an embedded supercharged diesel tank according to an embodiment of the present application. As shown in fig. 2, in this application scenario, first, a monitoring image of the diesel master tank 1 of the embedded supercharged diesel tank acquired by a camera is acquired (for example, C as illustrated in fig. 2); the acquired monitoring image is then input into a server (e.g., S as illustrated in fig. 2) that deploys a monitoring algorithm for the embedded pressurized diesel tank, wherein the server is capable of processing the monitoring image based on the monitoring algorithm for the embedded pressurized diesel tank to generate a classification result that indicates whether a leak warning prompt is generated.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
In one embodiment of the present application, fig. 3 is a block diagram of a monitoring system for an embedded pressurized diesel tank according to an embodiment of the present application. As shown in fig. 3, a monitoring system 100 of an embedded pressurized diesel tank according to an embodiment of the present application includes: the image acquisition module 110 is used for acquiring a monitoring image of the diesel master tank of the embedded type supercharged diesel tank acquired by the camera;
The image preprocessing module 120 is configured to perform image preprocessing on the monitoring image of the diesel tank to obtain a preprocessed monitoring image; the feature extraction module 130 is configured to pass the preprocessed monitoring image through a convolutional neural network model serving as a feature extractor to obtain a surface feature matrix of the diesel tank; the feature matrix dividing module 140 is configured to divide a feature matrix of the surface feature matrix of the diesel tank to obtain a plurality of local feature matrices; the context coding module 150 is configured to obtain a diesel tank surface global feature matrix through a context coder based on a converter after the plurality of local feature matrices are expanded into a plurality of local expansion feature vectors; and the early warning prompt generation module 160 is configured to pass the global feature matrix on the surface of the diesel tank through a classifier to obtain a classification result, where the classification result is used to indicate whether a leakage early warning prompt is generated.
Specifically, in the embodiment of the present application, the image obtaining module 110 is configured to obtain the monitoring image of the diesel tank of the embedded pressurized diesel tank, which is collected by the camera. In the operation process of the embedded supercharged diesel oil tank system, the embedded supercharged diesel oil tank is required to be subjected to leakage monitoring so as to ensure that the embedded supercharged diesel oil tank is stable in operation.
Aiming at the technical requirements, the technical conception of the application is as follows: the external surface image of the diesel oil master tank is acquired through the camera, and then the external surface image of the diesel oil master tank is subjected to image processing and analysis, so that the leakage detection is realized. Here, it should be understood that if the diesel tank leaks, and oil stains may occur on the outer surface, so that the image of the outer surface of the diesel tank changes, it may be determined whether the diesel tank leaks based on the change of the image end.
Specifically, in the technical scheme of the application, firstly, a monitoring image of a diesel master tank of an embedded type supercharged diesel tank collected by a camera is obtained. It will be appreciated that in the embedded pressurized diesel tank system, the diesel tank is an integral part of the diesel storage in the system, and therefore monitoring the diesel tank is very important. Correspondingly, the monitoring image of the diesel oil master tank is acquired by the camera, so that the condition of the surface of the diesel oil master tank can be observed in real time.
Specifically, in the embodiment of the present application, the image preprocessing module 120 is configured to perform image preprocessing on the monitoring image of the diesel tank to obtain a preprocessed monitoring image. In practical applications, the acquired image may contain various noises, such as uneven brightness, artifacts, shadows, etc., due to uncertainty of the camera environment and interference of other factors. These noises can affect the accuracy of subsequent computer vision and deep learning techniques, and therefore require pre-processing of the image to remove or reduce these disturbances. Accordingly, by preprocessing the monitored image, operations such as background noise elimination, edge enhancement, contrast enhancement and the like can be performed, so that the extracted features are more accurate and stable. For example, a filter may be employed to smooth the image to eliminate noise; histogram equalization may also be used to adjust the brightness and contrast of the image.
Wherein, the image preprocessing module 120 is configured to: and carrying out image preprocessing on the monitoring image of the diesel oil master tank by adopting a filter and histogram equalization to obtain the preprocessed monitoring image.
Specifically, in the embodiment of the present application, the feature extraction module 130 is configured to pass the preprocessed monitoring image through a convolutional neural network model serving as a feature extractor to obtain a diesel tank surface feature matrix. After image preprocessing, the preprocessed monitoring image is passed through a convolutional neural network model serving as a feature extractor to obtain a diesel tank surface feature matrix. Namely, the convolution neural network model with excellent performance in the field of image feature extraction is used for carrying out image local feature extraction based on convolution kernels on the preprocessed monitoring image so as to obtain the diesel oil master tank surface feature matrix.
Those of ordinary skill in the art will appreciate that the convolutional neural network is a deep learning model widely used in the field of computer vision that automatically learns feature representations from a large amount of data and efficiently processes two-dimensional data such as images. In the scheme, a pre-trained convolutional neural network model is used as a feature extractor, and convolution processing, pooling processing along the channel dimension and nonlinear activation processing are carried out on the extracted monitoring image to obtain the diesel oil master tank surface feature matrix.
Wherein, the feature extraction module 130 is configured to: and respectively carrying out convolution processing, pooling processing along a channel dimension and nonlinear activation processing on input data in forward transmission of layers by using each layer of the convolutional neural network model serving as a feature extractor to output the final layer of the convolutional neural network model serving as the feature extractor as the diesel tank surface feature matrix, wherein the input of the first layer of the convolutional neural network model serving as the feature extractor is the preprocessed monitoring image.
The convolutional neural network is an artificial neural network and has wide application in the fields of image recognition and the like. The convolutional neural network may include an input layer, a hidden layer, and an output layer, where the hidden layer may include a convolutional layer, a pooling layer, an activation layer, a full connection layer, etc., where the previous layer performs a corresponding operation according to input data, outputs an operation result to the next layer, and obtains a final result after the input initial data is subjected to a multi-layer operation.
The convolutional neural network model has excellent performance in the aspect of image local feature extraction by taking a convolutional kernel as a feature filtering factor, and has stronger feature extraction generalization capability and fitting capability compared with the traditional image feature extraction algorithm based on statistics or feature engineering.
Specifically, in the embodiment of the present application, the feature matrix dividing module 140 is configured to perform feature matrix division on the surface feature matrix of the diesel tank to obtain a plurality of local feature matrices. In order to capture long-range dependency information among various local features, in the technical scheme of the application, firstly, feature matrix division is carried out on the diesel oil master tank surface feature matrix to obtain a plurality of local feature matrices, the local feature matrices are unfolded into a plurality of local unfolded feature vectors, and then the local feature matrices are unfolded through a context encoder based on a converter to obtain the diesel oil master tank surface global feature matrix.
Specifically, the feature matrix division is performed on the surface feature matrix of the diesel oil master tank, so that a large and high-dimensional feature matrix is divided into a plurality of smaller local feature matrices, and preamble support is provided for context coding among subsequent local feature matrices.
Wherein, the feature matrix dividing module 140 is configured to: and uniformly dividing the surface feature matrix of the diesel oil mother tank to obtain a plurality of local feature matrices.
Specifically, in the embodiment of the present application, the context encoding module 150 is configured to obtain the diesel tank surface global feature matrix by using a context encoder based on a converter after expanding the plurality of local feature matrices into a plurality of locally expanded feature vectors. Then, the local feature matrices are expanded into local expansion feature vectors, and then the local expansion feature vectors are obtained through a context encoder based on a converter, wherein the context encoder based on the converter utilizes a self-attention mechanism in the concept of the converter to capture global context association information of the local expansion feature vectors relative to the local expansion feature vectors so as to obtain the local expansion feature vectors. And after the plurality of context local expansion feature vectors are obtained, the plurality of context local expansion feature vectors are arranged into the diesel oil master tank surface global feature matrix.
Fig. 4 is a block diagram of the context encoding module in the monitoring system of the embedded supercharged diesel tank according to the embodiment of the present application, as shown in fig. 4, the context encoding module 150 includes: a self-attention unit 151 for performing context semantic coding based on a self-attention mechanism on the plurality of local expansion feature vectors using the converter-based context encoder to obtain a plurality of context local expansion feature vectors; a gaussian regression factor calculation unit 152 for calculating a gaussian regression uncertainty factor for each of the plurality of contextual locally expanded feature vectors; a weighting unit 153, configured to respectively weight each of the local expansion feature vectors of the context with a gaussian regression uncertainty factor of the local expansion feature vector of the context as a weight to obtain a plurality of weighted local expansion feature vectors of the context; and a two-dimensional arrangement unit 154, configured to two-dimensionally arrange the plurality of weighted context local expansion feature vectors to obtain the diesel tank surface global feature matrix.
Specifically, in the technical scheme of the application, after the characteristic matrix of the surface of the diesel oil master tank is obtained through a convolution neural network model serving as a characteristic extractor and is divided by the characteristic matrix, gaussian distribution error uncertainty of characteristic distribution of each local characteristic matrix is caused, after context correlation coding of image semantic characteristics is carried out based on a context encoder of a converter, gaussian distribution error uncertainty of characteristic distribution of each characteristic vector in a plurality of context local expansion characteristic vectors is further carried out, and classification regression errors are caused in the global characteristic matrix of the surface of the diesel oil master tank obtained by two-dimensional arrangement of the plurality of context local expansion characteristic vectors, so that accuracy of classification results obtained by the classifier of the global characteristic matrix of the surface of the diesel oil master tank is affected.
Based on this, in the technical solution of the present application, a gaussian regression uncertainty factor of each of the plurality of context local expansion feature vectors is calculated, and is expressed as: calculating a gaussian regression uncertainty factor for each of the plurality of contextual locally-expanded feature vectors with an optimization formula; wherein, the optimization formula is:
，
Wherein,,is the length of the feature vector, +.>And->The mean and variance of feature sets at each location in the plurality of contextual locally expanded feature vectors, respectively, wherein +.>Is the +.f of the local expansion feature vector of the plurality of contexts>Characteristic value of individual position, and->To take the following measures2, a logarithmic function with a base>Is the gaussian regression uncertainty factor.
Here, for the agnostic regression of the global feature matrix of the diesel tank surface, which may be caused by the distribution uncertainty information of the integrated feature set of each of the plurality of context local expansion feature vectors, the scalar measurement of the statistical characteristics of the feature set is performed by using the mean value and the variance as the statistical quantization parameters, so that the normal distribution cognitive mode represented by the feature errors is expanded to an unknown distribution regression mode, and the migration learning based on natural distribution transfer on the feature set scale is realized, so that the global feature matrix of the diesel tank surface is obtained by weighting each context local expansion feature vector by the gaussian regression uncertainty factors and then two-dimensionally arranging the weighted context local expansion feature vectors, and the uncertainty correction based on self calibration of each context local expansion feature vector when the global feature matrix of the diesel tank surface is formed, so that the classification regression errors existing in the global feature matrix of the diesel tank surface are corrected, and the accuracy of the classification result obtained by the classifier is improved.
Specifically, in the embodiment of the present application, the early warning prompt generation module 160 is configured to pass the global feature matrix on the surface of the diesel tank through a classifier to obtain a classification result, where the classification result is used to indicate whether a leakage early warning prompt is generated. And finally, the global feature matrix on the surface of the diesel oil master tank passes through a classifier to obtain a classification result, wherein the classification result is used for indicating whether leakage early warning prompt is generated or not. The classifier is used for determining a class probability label of the global feature matrix of the surface of the diesel tank, and the class probability label is used for indicating whether leakage early warning prompt is generated or not. Thus, an intelligent monitoring scheme of the diesel fuel mother tank is constructed based on machine vision and intelligent image analysis and processing calculation.
Fig. 5 is a block diagram of the early warning prompt generation module in the monitoring system of the embedded supercharged diesel tank according to the embodiment of the present application, as shown in fig. 5, the early warning prompt generation module 160 includes: a matrix expansion unit 161, configured to expand the global feature matrix on the surface of the diesel tank into classification feature vectors according to row vectors or column vectors; a full-connection encoding unit 162, configured to perform full-connection encoding on the classification feature vector by using multiple full-connection layers of the classifier to obtain an encoded classification feature vector; and a classification unit 163, configured to pass the encoded classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
In summary, a monitoring system 100 of an embedded pressurized diesel tank according to an embodiment of the present application is illustrated, which acquires a monitoring image of a diesel master tank of the embedded pressurized diesel tank acquired by a camera; and (3) performing image processing and analysis on the outer surface image of the diesel oil master tank by adopting an artificial intelligence technology based on deep learning, and generating an early warning prompt whether leakage occurs or not based on the image processing and the analysis. Thus, the leakage monitoring can be intelligently performed on the diesel oil master tank so as to ensure the running stability of the embedded supercharged diesel oil tank.
As described above, the monitoring system 100 of an embedded type supercharged diesel tank according to the embodiment of the present application may be implemented in various terminal devices, for example, a server for monitoring an embedded type supercharged diesel tank, or the like. In one example, the embedded pressurized diesel tank monitoring system 100 according to embodiments of the present application may be integrated into the terminal device as a software module and/or hardware module. For example, the monitoring system 100 of the embedded pressurized diesel tank may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the monitoring system 100 of the embedded pressurized diesel tank may also be one of a plurality of hardware modules of the terminal device.
Alternatively, in another example, the monitoring system 100 of the embedded pressurized diesel tank and the terminal device may be separate devices, and the monitoring system 100 of the embedded pressurized diesel tank may be connected to the terminal device through a wired and/or wireless network and transmit the interaction information according to a agreed data format.
In one embodiment of the present application, fig. 6 is a flowchart of a method of monitoring an embedded pressurized diesel tank according to an embodiment of the present application. As shown in fig. 6, a method for monitoring an embedded supercharged diesel tank according to an embodiment of the present application includes: 210, acquiring a monitoring image of a diesel master tank of the embedded pressurized diesel tank acquired by a camera; 220, performing image preprocessing on the monitoring image of the diesel fuel mother tank to obtain a preprocessed monitoring image; 230, passing the preprocessed monitoring image through a convolutional neural network model serving as a feature extractor to obtain a diesel oil mother tank surface feature matrix; 240, performing feature matrix division on the diesel oil mother tank surface feature matrix to obtain a plurality of local feature matrices; 250, expanding the local feature matrixes into a plurality of local expansion feature vectors, and then obtaining a global feature matrix of the surface of the diesel tank through a context encoder based on a converter; and 260, passing the global feature matrix on the surface of the diesel tank through a classifier to obtain a classification result, wherein the classification result is used for indicating whether leakage early warning prompt is generated or not.
Fig. 7 is a schematic diagram of a system architecture of a monitoring method of an embedded supercharged diesel tank according to an embodiment of the present application. As shown in fig. 7, in the system architecture of the method for monitoring an embedded supercharged diesel tank, firstly, a monitoring image of a diesel master tank of the embedded supercharged diesel tank collected by a camera is obtained; then, carrying out image preprocessing on the monitoring image of the diesel oil master tank to obtain a preprocessed monitoring image; then, the preprocessed monitoring image passes through a convolutional neural network model serving as a feature extractor to obtain a diesel oil mother tank surface feature matrix; then, carrying out feature matrix division on the surface feature matrix of the diesel oil master tank to obtain a plurality of local feature matrices; then, the local feature matrixes are unfolded to be a plurality of local unfolded feature vectors, and the local feature vectors are passed through a context encoder based on a converter to obtain a global feature matrix of the surface of the diesel tank; and finally, the global feature matrix on the surface of the diesel oil master tank is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether leakage early warning prompt is generated or not.
In a specific example, in the above monitoring method of an embedded supercharged diesel tank, the image preprocessing is performed on the monitoring image of the diesel tank to obtain a preprocessed monitoring image, including: and carrying out image preprocessing on the monitoring image of the diesel oil master tank by adopting a filter and histogram equalization to obtain the preprocessed monitoring image.
In a specific example, in the above monitoring method of an embedded supercharged diesel tank, passing the preprocessed monitoring image through a convolutional neural network model as a feature extractor to obtain a diesel master tank surface feature matrix, including: and respectively carrying out convolution processing, pooling processing along a channel dimension and nonlinear activation processing on input data in forward transmission of layers by using each layer of the convolutional neural network model serving as a feature extractor to output the final layer of the convolutional neural network model serving as the feature extractor as the diesel tank surface feature matrix, wherein the input of the first layer of the convolutional neural network model serving as the feature extractor is the preprocessed monitoring image.
In a specific example, in the above monitoring method of an embedded supercharged diesel tank, performing feature matrix division on the surface feature matrix of the diesel parent tank to obtain a plurality of local feature matrices, including: and uniformly dividing the surface feature matrix of the diesel oil mother tank to obtain a plurality of local feature matrices.
In a specific example, in the above monitoring method of an embedded supercharged diesel tank, the developing the plurality of local feature matrices into a plurality of local developed feature vectors and then obtaining a global feature matrix of a diesel mother tank surface by a context encoder based on a converter includes: performing context semantic coding based on a self-attention mechanism on the plurality of local expansion feature vectors by using the context encoder based on the converter to obtain a plurality of context local expansion feature vectors; calculating a gaussian regression uncertainty factor for each of the plurality of contextual locally-expanded feature vectors; taking the Gaussian regression uncertainty factor of each context local expansion feature vector as a weight, and respectively weighting each context local expansion feature vector to obtain a plurality of weighted context local expansion feature vectors; and performing two-dimensional arrangement on the weighted context local expansion feature vectors to obtain the diesel oil mother tank surface global feature matrix.
In a specific example, in the above-mentioned monitoring method of an embedded supercharged diesel tank, calculating a gaussian regression uncertainty factor of each of the plurality of contextual locally-developed feature vectors includes: calculating a gaussian regression uncertainty factor for each of the plurality of contextual locally-expanded feature vectors with an optimization formula; wherein, the optimization formula is:
，
wherein,,is the length of the feature vector, +.>And->The mean and variance of feature sets at each location in the plurality of contextual locally expanded feature vectors, respectively, wherein +.>Is the +.f of the local expansion feature vector of the plurality of contexts>Characteristic value of individual position, and->As a logarithmic function with base 2 +.>Is the gaussian regression uncertainty factor.
In a specific example, in the above monitoring method of an embedded supercharged diesel tank, the method includes passing the global feature matrix on the surface of the diesel tank through a classifier to obtain a classification result, where the classification result is used to indicate whether a leakage early warning prompt is generated, and the method includes: expanding the global feature matrix on the surface of the diesel oil mother tank into classification feature vectors according to row vectors or column vectors; performing full-connection coding on the classification feature vectors by using a plurality of full-connection layers of the classifier to obtain coded classification feature vectors; and passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
It will be appreciated by those skilled in the art that the specific operation of the steps in the above-described monitoring method of the embedded supercharged diesel tank has been described in detail in the above description of the monitoring system of the embedded supercharged diesel tank with reference to fig. 1 to 5, and thus, repeated descriptions thereof will be omitted.
The present application also provides a computer program product comprising instructions which, when executed, cause an apparatus to perform operations corresponding to the above-described methods.
In one embodiment of the present application, there is also provided a computer readable storage medium storing a computer program for executing the above-described method.
It should be appreciated that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the forms of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects may be utilized. Furthermore, the computer program product may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
Methods, systems, and computer program products of embodiments of the present application are described in terms of flow diagrams and/or block diagrams. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer program, such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The basic principles of the present application have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not limiting, and these advantages, benefits, effects, etc. are not to be considered as necessarily possessed by the various embodiments of the present application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not intended to be limited to the details disclosed herein as such.
The block diagrams of the devices, apparatuses, devices, systems referred to in this application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.
It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent to the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.
Claims (8)
1. The utility model provides a monitoring system of embedded supercharged diesel tank which characterized in that includes:
the image acquisition module is used for acquiring a monitoring image of the diesel master tank of the embedded type supercharged diesel tank acquired by the camera;
the image preprocessing module is used for preprocessing the image of the monitoring image of the diesel oil master tank to obtain a preprocessed monitoring image;
the feature extraction module is used for enabling the preprocessed monitoring image to pass through a convolutional neural network model serving as a feature extractor to obtain a diesel oil mother tank surface feature matrix;
the characteristic matrix dividing module is used for dividing the characteristic matrix of the surface of the diesel oil mother tank to obtain a plurality of local characteristic matrixes;
the context coding module is used for expanding the local feature matrixes into a plurality of local expansion feature vectors and then obtaining a global feature matrix on the surface of the diesel oil mother tank through a context coder based on a converter; and
The early warning prompt generation module is used for enabling the global feature matrix on the surface of the diesel oil mother tank to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether leakage early warning prompts are generated or not;
wherein the context encoding module comprises:
a self-attention unit for performing context semantic coding based on a self-attention mechanism on the plurality of local expansion feature vectors by using the context encoder based on the converter to obtain a plurality of context local expansion feature vectors;
a gaussian regression factor calculation unit for calculating a gaussian regression uncertainty factor for each of the plurality of contextual locally-expanded feature vectors;
the weighting unit is used for respectively weighting each context local expansion feature vector by taking the Gaussian regression uncertainty factor of each context local expansion feature vector as a weight so as to obtain a plurality of weighted context local expansion feature vectors; and
the two-dimensional arrangement unit is used for carrying out two-dimensional arrangement on the plurality of weighted context local expansion feature vectors to obtain the diesel oil mother tank surface global feature matrix;
Wherein, the Gaussian regression factor calculation unit is used for: calculating a gaussian regression uncertainty factor for each of the plurality of contextual locally-expanded feature vectors with an optimization formula;
wherein, the optimization formula is:
，
wherein,,is the length of the feature vector, +.>And->The mean and variance of feature sets at each location in the plurality of contextual locally expanded feature vectors, respectively, wherein +.>Is the +.f of the local expansion feature vector of the plurality of contexts>Characteristic value of individual position, and->As a logarithmic function with base 2 +.>Is the gaussian regression uncertainty factor.
2. The embedded pressurized diesel tank monitoring system of claim 1, wherein the image preprocessing module is configured to: and carrying out image preprocessing on the monitoring image of the diesel oil master tank by adopting a filter and histogram equalization to obtain the preprocessed monitoring image.
3. The embedded pressurized diesel tank monitoring system of claim 2, wherein the feature extraction module is configured to: and respectively carrying out convolution processing, pooling processing along a channel dimension and nonlinear activation processing on input data in forward transmission of layers by using each layer of the convolutional neural network model serving as a feature extractor to output the final layer of the convolutional neural network model serving as the feature extractor as the diesel tank surface feature matrix, wherein the input of the first layer of the convolutional neural network model serving as the feature extractor is the preprocessed monitoring image.
4. The embedded pressurized diesel tank monitoring system of claim 3, wherein the feature matrix partitioning module is configured to: and uniformly dividing the surface feature matrix of the diesel oil mother tank to obtain a plurality of local feature matrices.
5. The monitoring system of an embedded pressurized diesel tank of claim 4, wherein the early warning hint generation module comprises:
the matrix unfolding unit is used for unfolding the global feature matrix on the surface of the diesel oil mother tank into classification feature vectors according to row vectors or column vectors;
the full-connection coding unit is used for carrying out full-connection coding on the classification characteristic vectors by using a plurality of full-connection layers of the classifier so as to obtain coded classification characteristic vectors; and
and the classification unit is used for passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
6. The monitoring method of the embedded supercharged diesel tank is characterized by comprising the following steps of:
acquiring a monitoring image of a diesel master tank of an embedded pressurized diesel tank acquired by a camera;
performing image preprocessing on the monitoring image of the diesel oil master tank to obtain a preprocessed monitoring image;
The preprocessed monitoring image passes through a convolutional neural network model serving as a feature extractor to obtain a diesel oil mother tank surface feature matrix;
performing feature matrix division on the surface feature matrix of the diesel oil mother tank to obtain a plurality of local feature matrices;
expanding the local feature matrixes into a plurality of local expansion feature vectors, and then passing through a context encoder based on a converter to obtain a global feature matrix on the surface of the diesel tank; and
the global feature matrix on the surface of the diesel oil mother tank is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether leakage early warning prompt is generated;
the method for obtaining the global feature matrix of the diesel tank surface through a context encoder based on a converter after expanding the local feature matrices into a plurality of local expansion feature vectors comprises the following steps:
performing context semantic coding based on a self-attention mechanism on the plurality of local expansion feature vectors by using the context encoder based on the converter to obtain a plurality of context local expansion feature vectors;
calculating a gaussian regression uncertainty factor for each of the plurality of contextual locally-expanded feature vectors;
Taking the Gaussian regression uncertainty factor of each context local expansion feature vector as a weight, and respectively weighting each context local expansion feature vector to obtain a plurality of weighted context local expansion feature vectors; and
performing two-dimensional arrangement on the plurality of weighted context local expansion feature vectors to obtain a global feature matrix of the diesel oil mother tank surface;
wherein calculating a gaussian regression uncertainty factor for each of the plurality of contextual locally-expanded feature vectors comprises: calculating a gaussian regression uncertainty factor for each of the plurality of contextual locally-expanded feature vectors with an optimization formula;
wherein, the optimization formula is:
，
wherein,,is the length of the feature vector, +.>And->The mean and variance of feature sets at each location in the plurality of contextual locally expanded feature vectors, respectively, wherein +.>Is the +.f of the local expansion feature vector of the plurality of contexts>Characteristic value of individual position, and->As a logarithmic function with base 2 +.>Is the gaussian regression uncertainty factor.
7. The method for monitoring an embedded supercharged diesel tank of claim 6, wherein image preprocessing is performed on the monitoring image of the diesel parent tank to obtain a preprocessed monitoring image, comprising: and carrying out image preprocessing on the monitoring image of the diesel oil master tank by adopting a filter and histogram equalization to obtain the preprocessed monitoring image.
8. The method for monitoring the embedded supercharged diesel tank according to claim 7, wherein the step of passing the preprocessed monitoring image through a convolutional neural network model as a feature extractor to obtain a diesel master tank surface feature matrix comprises the steps of: and respectively carrying out convolution processing, pooling processing along a channel dimension and nonlinear activation processing on input data in forward transmission of layers by using each layer of the convolutional neural network model serving as a feature extractor to output the final layer of the convolutional neural network model serving as the feature extractor as the diesel tank surface feature matrix, wherein the input of the first layer of the convolutional neural network model serving as the feature extractor is the preprocessed monitoring image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310547149.0A CN116259012B (en) | 2023-05-16 | 2023-05-16 | Monitoring system and method for embedded supercharged diesel tank |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310547149.0A CN116259012B (en) | 2023-05-16 | 2023-05-16 | Monitoring system and method for embedded supercharged diesel tank |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116259012A CN116259012A (en) | 2023-06-13 |
| CN116259012B true CN116259012B (en) | 2023-07-28 |
Family
ID=86681046
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310547149.0A Active CN116259012B (en) | 2023-05-16 | 2023-05-16 | Monitoring system and method for embedded supercharged diesel tank |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116259012B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116739868B (en) * | 2023-07-05 | 2024-04-23 | 浙江星宸环境建设有限公司 | Afforestation management system and method based on artificial intelligence |
| CN116630081B (en) * | 2023-07-25 | 2023-09-29 | 新疆华屹能源发展有限公司 | Front-mounted energy-increasing viscosity-reducing yield-increasing method for nitrogen in oil production well |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115789517A (en) * | 2023-01-04 | 2023-03-14 | 河南纳宇新材料有限公司 | Solid-state hydrogen storage tank with leak testing function |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109255797B (en) * | 2017-07-14 | 2021-08-06 | 富士通株式会社 | Image processing device and method, and electronic device |
| CN110136210B (en) * | 2018-02-07 | 2023-06-09 | 中国石油化工股份有限公司 | Intelligent early warning system for leakage of monitoring device |
| KR102067242B1 (en) * | 2018-05-23 | 2020-01-16 | 한국해양과학기술원 | Method for detecting oil spills on satellite sar images using artificial neural network |
| CN109325520B (en) * | 2018-08-24 | 2021-06-29 | 北京航空航天大学 | Method, device and system for checking petroleum leakage |
| DE102020212502A1 (en) * | 2020-10-02 | 2022-04-07 | Robert Bosch Gesellschaft mit beschränkter Haftung | BAYESAN CONTEXT AGGREGATION FOR NEURAL PROCESSES |
| CN115909260A (en) * | 2022-09-28 | 2023-04-04 | 华能伊敏煤电有限责任公司 | Method and system for early warning of workplace intrusion based on machine vision |
| CN115496740B (en) * | 2022-10-10 | 2023-05-16 | 湖北华鑫光电有限公司 | Lens defect detection method and system based on convolutional neural network |
| CN116052254A (en) * | 2023-01-19 | 2023-05-02 | 西安邮电大学 | Visual continuous emotion recognition method based on extended Kalman filtering neural network |
-
2023
- 2023-05-16 CN CN202310547149.0A patent/CN116259012B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115789517A (en) * | 2023-01-04 | 2023-03-14 | 河南纳宇新材料有限公司 | Solid-state hydrogen storage tank with leak testing function |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116259012A (en) | 2023-06-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116259012B (en) | Monitoring system and method for embedded supercharged diesel tank | |
| CN108008948B (en) | Multiplexing device, multiplexing method and processing device for instruction generation process | |
| CN116092701B (en) | Control system and method based on health data analysis management platform | |
| CN113076957A (en) | RGB-D image saliency target detection method based on cross-modal feature fusion | |
| CN116086790A (en) | Performance detection method and system for high-pressure valve of hydrogen fuel cell | |
| CN113095414A (en) | Indicator diagram identification method based on convolutional neural network and support vector machine | |
| CN116796269A (en) | Management method and system for Internet of things equipment | |
| CN114821328A (en) | Electric power image processing method and device based on complete learning | |
| CN115937689A (en) | Agricultural pest intelligent identification and monitoring technology | |
| CN115951883A (en) | Service component management system and method of distributed micro-service architecture | |
| CN117498321A (en) | Distributed photovoltaic output prediction method, system and storage medium | |
| CN110852522B (en) | Short-term power load prediction method and system | |
| CN115564092A (en) | Short-time wind power prediction system and method for wind power plant | |
| CN117372879B (en) | Lightweight remote sensing image change detection method and system based on self-supervision enhancement | |
| CN111578154B (en) | LSDR-JMI-based water supply network multi-leakage pressure sensor optimal arrangement method | |
| CN116721071B (en) | Industrial product surface defect detection method and device based on weak supervision | |
| CN116012688B (en) | Image enhancement method for urban management evaluation system | |
| CN116091873B (en) | Image generation method, device, electronic equipment and storage medium | |
| CN115099129B (en) | Natural gas well yield prediction method based on input characteristic error correction | |
| CN115979548B (en) | Method, system, electronic device and storage medium for diagnosing leakage of hydrogen system for vehicle | |
| CN116653861A (en) | Defogging control system and defogging control method for automobile rearview mirror | |
| He et al. | Rolling bearing fault diagnosis based on meta-learning with few-shot samples | |
| CN116977840A (en) | Marine organism target detection method, system, storage medium and equipment | |
| CN116639794A (en) | Medical wastewater disinfection treatment system and treatment method | |
| CN116824330A (en) | Small sample cross-domain target detection method based on deep learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |