CN114724246A - Dangerous behavior identification method and device - Google Patents
Dangerous behavior identification method and device Download PDFInfo
- Publication number
- CN114724246A CN114724246A CN202210375655.1A CN202210375655A CN114724246A CN 114724246 A CN114724246 A CN 114724246A CN 202210375655 A CN202210375655 A CN 202210375655A CN 114724246 A CN114724246 A CN 114724246A
- Authority
- CN
- China
- Prior art keywords
- real
- dangerous behavior
- time image
- dangerous
- behavior recognition
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000012549 training Methods 0.000 claims abstract description 20
- 230000006399 behavior Effects 0.000 claims description 148
- 238000004590 computer program Methods 0.000 claims description 25
- 230000007246 mechanism Effects 0.000 claims description 14
- 238000007781 pre-processing Methods 0.000 claims description 11
- 238000003860 storage Methods 0.000 claims description 11
- 231100001261 hazardous Toxicity 0.000 claims description 5
- 238000001514 detection method Methods 0.000 description 15
- 238000010586 diagram Methods 0.000 description 13
- 238000012545 processing Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 5
- 230000003044 adaptive effect Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000004931 aggregating effect Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 235000019580 granularity Nutrition 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a dangerous behavior identification method and a dangerous behavior identification device, wherein the method comprises the following steps: acquiring a real-time image of a target area; inputting the real-time image into a dangerous behavior recognition model, and outputting a dangerous behavior recognition result of the real-time image; the dangerous behavior recognition model is obtained by training according to a vehicle sample and a person sample and labels corresponding to the vehicle sample and the person sample based on a yolov5 algorithm. According to the dangerous behavior identification method and device disclosed by the invention, the dangerous behavior identification result of the real-time image is output by acquiring the real-time image of the target area, inputting the real-time image into the dangerous behavior identification model based on the yolov5 algorithm, and more efficient, timely and accurate dangerous behavior identification can be carried out.
Description
Technical Field
The present invention relates to the field of computers, and more particularly, to a method and an apparatus for identifying dangerous behaviors.
Background
At present, dangerous behaviors in a specific area such as a battlefield are generally visually recognized by a human. Therefore, the method has the defects of low efficiency, untimely discovery, limited discovery distance, poor accuracy and the like.
Disclosure of Invention
The invention aims to provide a dangerous behavior identification method and a dangerous behavior identification device, which can identify dangerous behaviors more efficiently, timely and accurately.
In order to achieve the above object, the present invention provides a dangerous behavior recognition method, including:
acquiring a real-time image of a target area;
inputting the real-time image into a dangerous behavior recognition model, and outputting a dangerous behavior recognition result of the real-time image;
the dangerous behavior recognition model is obtained by training according to a vehicle sample and a person sample and labels corresponding to the vehicle sample and the person sample based on a yolov5 algorithm.
In an embodiment of the present invention, after the inputting the real-time image into a dangerous behavior recognition model and outputting a dangerous behavior recognition result of the real-time image, the method further includes:
and sending an instruction corresponding to the target behavior to an execution mechanism under the condition that the dangerous behavior identification result is the target behavior, so that the execution mechanism executes the operation corresponding to the target behavior.
In an embodiment of the present invention, the acquiring a real-time image of a target area, inputting the real-time image into a dangerous behavior recognition model, and outputting a dangerous behavior recognition result of the real-time image, further includes:
and preprocessing the real-time image.
In one embodiment of the present invention, a dangerous behavior recognition apparatus includes:
the image acquisition module is used for acquiring a real-time image of the target area;
the target recognition module is used for inputting the real-time image into a dangerous behavior recognition model and outputting a dangerous behavior recognition result of the real-time image;
the dangerous behavior recognition model is obtained by training according to a vehicle sample and a person sample and labels corresponding to the vehicle sample and the person sample based on a yolov5 algorithm.
In an embodiment of the present invention, the dangerous behavior recognition apparatus further includes:
and the instruction sending module is used for sending an instruction corresponding to the target behavior to an execution mechanism under the condition that the dangerous behavior identification result is the target behavior, so that the execution mechanism executes the operation corresponding to the target behavior.
In an embodiment of the present invention, the dangerous behavior recognition apparatus further includes:
and the image preprocessing module is used for preprocessing the real-time image.
In an embodiment of the present invention, an electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the steps of the method for identifying dangerous behavior based on smart contracts according to any one of the above methods.
In an embodiment of the present invention, a non-transitory computer readable storage medium has stored thereon a computer program which, when being executed by a processor, implements the steps of the intelligent contract-based dangerous behavior identification method as described in any one of the above.
In an embodiment of the present invention, a computer program product includes a computer program, and the computer program realizes the steps of the dangerous behavior identification method based on the intelligent contract according to any one of the above mentioned methods when being executed by a processor.
Compared with the prior art, according to the dangerous behavior recognition method and device, the dangerous behavior recognition can be carried out more efficiently, timely and accurately by acquiring the real-time image of the target area, inputting the real-time image into the dangerous behavior recognition model based on the yolov5 algorithm, and outputting the dangerous behavior recognition result of the real-time image.
Drawings
FIG. 1 is a flow diagram illustrating a method for identifying dangerous behavior according to one embodiment of the present invention;
FIG. 2 is a second flowchart illustrating a dangerous behavior recognition method according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a dangerous behavior recognition apparatus according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the invention.
Detailed Description
The following detailed description of the present invention is provided in conjunction with the accompanying drawings, but it should be understood that the scope of the present invention is not limited to the specific embodiments.
Throughout the specification and claims, unless explicitly stated otherwise, the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element or component but not the exclusion of any other element or component.
As shown in fig. 1 to 4, the method and apparatus for recognizing dangerous behaviors according to the preferred embodiment of the present invention may be implemented in the following manner.
Fig. 1 is a flowchart illustrating a dangerous behavior recognition method according to an embodiment of the present invention.
As shown in fig. 1, the method may include the steps of:
102, inputting the real-time image into a dangerous behavior recognition model, and outputting a dangerous behavior recognition result of the real-time image; the dangerous behavior recognition model is obtained by training according to the vehicle sample, the person sample and labels corresponding to the vehicle sample and the person sample based on the yolov5 algorithm.
Specifically, the dangerous behavior identification method provided by the embodiment of the invention is a dangerous behavior identification method based on yolov5(You Only Look Once v5) algorithm. The dangerous behavior identification method can be used for identifying dangerous behaviors in battlefields, specific places (such as key safety protection units) and the like.
The method is based on the deep learning technology, has the advantages of flexible structure, automatic feature extraction, high detection precision, high detection speed and the like, and can more flexibly deal with emergency situations and improve the hitting capability in military confrontation.
In one embodiment, a mode that an unmanned aerial vehicle carries a camera can be adopted to acquire image data in real time and acquire a real-time image of a target area.
In one embodiment, the drone may be provided with a three-light pod, the drone cruising alert, the pod acquiring image data in real time, the acquired image being transmitted to the background command system in real time. A background director system (which may be referred to simply as "background") may perform step 102.
In one embodiment, the unmanned aerial vehicle can access ad hoc network equipment, guarantee remote data transmission, and transmit a real-time image of a target area to a background in an ad hoc network mode. The background can input the real-time image into the dangerous behavior recognition model for prediction to obtain a dangerous behavior recognition result.
In one embodiment, in the case that dangerous behaviors exist, the dangerous behavior recognition result may be a feature map in which dangerous behavior occurrence regions are marked in the real-time image.
The dangerous behavior may be a plurality of behaviors that are predetermined to cause damage to the target area.
In one embodiment, an execution subject of the dangerous behavior recognition method provided by the embodiment of the present invention may be a target recognition system (which may also be referred to as a dangerous behavior recognition apparatus). The target recognition system comprises three parts of image acquisition, target recognition and training system. The implementation of the embodiment of the invention mainly needs to develop two parts of a target recognition system and a training system of the target recognition system. The target identification part can receive the image data transmitted by the image acquisition part, synchronize with pose data of a camera carried by the unmanned aerial vehicle, perform target detection and positioning on the image data, and output results according to an interface protocol. The training system part can provide a dangerous behavior recognition model used by a machine learning algorithm for the target recognition part, and in the operation and maintenance process, a large number of samples are input into the training system to continuously perform machine learning, so that the recognition rate is improved.
The target recognition system can automatically detect and recognize targets such as vehicles, people and the like shot by the unmanned aerial vehicle at the overlooking angle according to the implementation image returned by the unmanned aerial vehicle-mounted camera through recognition algorithm design and training of massive vehicle and person samples. Optionally, the training samples and the recognition algorithm models can be continuously enriched at the later stage, and the recognition of other specific targets is expanded.
The method is characterized in that an unmanned aerial vehicle-mounted camera is used for collecting image information of a target area, and then a background-loaded computer is used for automatically detecting and identifying a target of an imaging target area, so that dangerous behaviors can be quickly, timely and efficiently detected. The unmanned platform is combined with a good computer vision technology to analyze the acquired image information, so that the effect of quickly detecting and identifying the target can be achieved.
The target detection and identification algorithm may mainly use a YOLO target detection algorithm. And when the YOLO target detection is calculated, the detection scale can be increased by reducing the number of convolution layers and dimensionality and combining the idea of a characteristic pyramid, so that the aim of improving the detection precision is fulfilled. The problem that a large number of images cannot be directly processed can be solved by combining a general processing framework of an image target detection algorithm based on deep learning.
The YOLO algorithm has high recognition accuracy and recognition speed, is a fast and compact open source object detection model, has vigorous vitality all the time in the YOLO family, extends from YOLO V1 to V5, continues for five generations nowadays, and is always used as one of the preferred frames of object detection by computer vision engineers with continuous innovation and perfection. The YOLOv5 algorithm has better performance compared with previous YOLO algorithms, the framework is lighter, and the small object recognition is greatly improved.
The YOLO v5 algorithm is an algorithm improved on the basis of the YOLO v3 algorithm. YOLO v5, like other YOLO models, is composed of three main components in its network structure, namely backhaul, cock and Prediction. The backhaul is used for aggregating and forming a convolutional neural network of image characteristics on different fine image granularities; the hack is used in a series of network layers that blend and combine image features and passes the image features to a prediction layer. (typically FPN or PANET); the Prediction is used for predicting image features, generating a bounding box and predicting a category. The most important modules CBL, Focus, CSP, SPP and PANET are used in these three components.
The CBL module structure is shown as a convolution structure of 1x1, and the CSP structure is composed of a CBL structure and a residual structure.
YOLO v5 defaults to 3x640x640 input, the Focus layer is used to copy it into four copies, then the four pictures are cut into four 3x320x320 slices by slicing operation, then the four slices are connected in depth by concat, the output is 12x320x320, then the 32x320x convolution layer is generated by convolution kernel number 32, and finally the result is input to the next convolution layer through batch _ norm and leaky _ relu.
SPP is a Spatial pyramid layer (Spatial posing layer), the input is 512x20x20, the output is 256x20x20 after passing through the convolution layer of 1x1, then the downsampling is carried out by parallel Maxpool of three different kernel _ sizes (5,9,13), note that all padding of max _ pool is same, so the result can be spliced and added with the initial feature to output 1024x20x20, and finally the result is restored to 512x20x20 by using a convolution kernel of 512.
The neutral Network part uses the PANET Network structure, the PAN structure is from the paper Aggregation Network, which is intended to be used in the task of Instance Segmentation (Instance Segmentation).
The feature extractor of the network adopts a new FPN structure of an enhanced Bottom-Up (Bottom Up) path, and improves the propagation of low-level features (part a). Each stage of the third pass takes as input the feature maps of the previous stage and processes them with 3x3 convolutional layers. The outputs are added by cross-connects to the same stage profiles of the top-down path, which provide information for the next stage (part b). While recovering the corrupted information path between each candidate region and all feature levels using Adaptive feature pooling (Adaptive feature posing), each candidate region on each feature level is aggregated to avoid being arbitrarily assigned (part c).
YOLO V5 mirrors a modified version of the PANET structure of YOLO V4-PANET typically uses adaptive feature pools to add adjacent layers together for mask prediction. However, this method is somewhat cumbersome when PANET is used in YOLO v4, and therefore, the authors of YOLO v4 did not add neighboring layers using adaptive feature pools, but performed Concat operations on them, thereby improving the accuracy of the prediction.
According to the embodiment of the invention, the real-time image of the target area is acquired, the real-time image is input into the dangerous behavior recognition model based on the yolov5 algorithm, the dangerous behavior recognition result of the real-time image is output, and the dangerous behavior recognition can be carried out more efficiently, timely and accurately.
In one embodiment, after inputting the real-time image into the dangerous behavior recognition model and outputting the dangerous behavior recognition result of the real-time image, the method further includes: and when the dangerous behavior identification result is the target behavior, sending an instruction corresponding to the target behavior to the execution mechanism so as to enable the execution mechanism to execute the operation corresponding to the target behavior.
Specifically, after the background receives the image and detects the dangerous behavior, a specified action or a striking behavior can be sent out.
The designated action or the striking behavior is sent out by sending an instruction corresponding to the target behavior to the execution mechanism. The execution mechanism can execute the operation corresponding to the target behavior in response to the instruction, and process the target behavior.
The target behavior may be one or more preset dangerous behaviors.
For example, the processing manner corresponding to the target behavior may be as shown in table 1.
Table 1 list of processing modes corresponding to target behaviors
In one embodiment, between acquiring a real-time image of a target area, inputting the real-time image into a dangerous behavior recognition model, and outputting a dangerous behavior recognition result of the real-time image, the method further includes: and preprocessing the real-time image.
Specifically, the target detection algorithm based on deep learning cannot directly process a large number of images, the images need to be cut and then detected, and a processing framework is provided for the images.
In the training stage, firstly, an image is cut into small image blocks with a certain size, and a large-size image label is changed along with the small image blocks; then sending the cut image small blocks and the labels into a designed deep learning network for training; during training, the difference value between the correct label information and the predicted label information is used as an error, network parameters are iteratively optimized, and finally a trained network model (namely a dangerous behavior recognition model) is stored.
In the detection stage, the test image is cut into small images; then, sending the small images into a network model trained in a training stage, and eliminating repeated positioning windows through an NMS post-processing algorithm to obtain detection results of the small images; and finally integrating the results to obtain the detection result of the original image.
Fig. 2 is a second flowchart of a dangerous behavior recognition method according to an embodiment of the invention.
As shown in fig. 2, the method may include the steps of:
and step 201, normally lifting the unmanned aerial vehicle.
Step 202, image acquisition is started.
The camera carried by the unmanned aerial vehicle can start to acquire images in real time.
And step 203, acquiring a target image.
The image of the target area captured by the camera may be taken as the target image.
And step 204, preprocessing the image.
The target image may be subjected to preprocessing operations such as cropping to a preset size.
And step 205, target identification.
And taking the dangerous behaviors as targets, and identifying whether the dangerous behaviors exist in the preprocessed image.
And step 206, judging whether the requirements are met.
If yes, ending; if not, execution may return to step 203. The requirement may be preset according to actual requirements, which is not specifically limited in the embodiment of the present invention.
In the following, the dangerous behavior recognition apparatus provided by the present invention is described, and the dangerous behavior recognition apparatus described below and the dangerous behavior recognition method described above may be referred to in correspondence with each other.
Fig. 3 is a schematic structural diagram of a dangerous behavior recognition device provided by the present invention. Based on the content of any one of the above embodiments, as shown in fig. 3, the apparatus includes an image capturing module 301 and an object identifying module 302, wherein:
the image acquisition module 301 is used for acquiring a real-time image of a target area;
the target identification module 302 is used for inputting the real-time image into the dangerous behavior identification model and outputting a dangerous behavior identification result of the real-time image;
the dangerous behavior recognition model is obtained by training according to the vehicle sample, the character sample and labels corresponding to the vehicle sample and the character sample based on the yolov5 algorithm.
Specifically, the image acquisition module 301 and the object recognition module 302 are electrically connected.
In one embodiment, the dangerous behavior recognition apparatus may further include:
and the instruction sending module is used for sending an instruction corresponding to the target behavior to the execution mechanism under the condition that the dangerous behavior identification result is the target behavior, so that the execution mechanism executes the operation corresponding to the target behavior.
In one embodiment, the dangerous behavior recognition apparatus may further include:
and the image preprocessing module is used for preprocessing the real-time image.
The dangerous behavior recognition device provided in the embodiment of the present invention is configured to execute the dangerous behavior recognition method according to the present invention, and an implementation manner of the dangerous behavior recognition device is consistent with an implementation manner of the dangerous behavior recognition method provided in the present invention, and may achieve the same beneficial effects, and details are not repeated here.
The dangerous behavior recognition device is used for the dangerous behavior recognition method of the foregoing embodiments. Therefore, the descriptions and definitions in the dangerous behavior recognition methods in the foregoing embodiments may be used for understanding the execution modules in the embodiments of the present invention.
According to the embodiment of the invention, the real-time image of the target area is acquired, the real-time image is input into the dangerous behavior recognition model based on the yolov5 algorithm, the dangerous behavior recognition result of the real-time image is output, and the dangerous behavior recognition can be carried out more efficiently, timely and accurately.
Fig. 4 illustrates a physical structure diagram of an electronic device, which may include, as shown in fig. 4: a processor (processor)410, a communication Interface 420, a memory (memory)430 and a communication bus 440, wherein the processor 410, the communication Interface 420 and the memory 430 are communicated with each other via the communication bus 440. Processor 410 may call logic instructions in memory 430 to perform a method of hazardous behavior identification, the method comprising: acquiring a real-time image of a target area; inputting the real-time image into a dangerous behavior recognition model, and outputting a dangerous behavior recognition result of the real-time image; the dangerous behavior recognition model is obtained by training according to a vehicle sample and a person sample and labels corresponding to the vehicle sample and the person sample based on a yolov5 algorithm.
In addition, the logic instructions in the memory 430 may be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk, and various media capable of storing program codes.
The processor 410 in the electronic device provided in the embodiment of the present application may call the logic instruction in the memory 430, and an implementation manner of the processor is consistent with an implementation manner of the method for identifying a dangerous behavior provided in the present application, and the same beneficial effects may be achieved, and details are not described here.
In another aspect, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, when the program instructions are executed by a computer, the computer being capable of executing the dangerous behavior identification method provided by the above methods, the method including: acquiring a real-time image of a target area; inputting the real-time image into a dangerous behavior recognition model, and outputting a dangerous behavior recognition result of the real-time image; the dangerous behavior recognition model is obtained by training according to a vehicle sample and a person sample and labels corresponding to the vehicle sample and the person sample based on a yolov5 algorithm.
When executed, the computer program product provided in the embodiment of the present application implements the above method for identifying dangerous behaviors, and the specific implementation manner of the method is consistent with the implementation manner described in the embodiment of the foregoing method, and the same beneficial effects can be achieved, and details are not repeated here.
In yet another aspect, the present invention also provides a non-transitory computer-readable storage medium having stored thereon a computer program, which when executed by a processor is implemented to perform the above-provided hazardous behavior identification method, the method comprising: acquiring a real-time image of a target area; inputting the real-time image into a dangerous behavior recognition model, and outputting a dangerous behavior recognition result of the real-time image; the dangerous behavior recognition model is obtained by training according to a vehicle sample and a person sample and labels corresponding to the vehicle sample and the person sample based on a yolov5 algorithm.
When a computer program stored on a non-transitory computer-readable storage medium provided in the embodiments of the present application is executed, the method for identifying a dangerous behavior is implemented, and a specific implementation manner of the method is consistent with the implementation manner described in the embodiments of the method, and the same beneficial effects can be achieved, and details are not repeated herein.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application 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, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams 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 implemented process 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 foregoing descriptions of specific exemplary embodiments of the present invention have been presented for purposes of illustration and description. It is not intended to limit the invention to the precise form disclosed, and obviously many modifications and variations are possible in light of the above teaching. The exemplary embodiments were chosen and described in order to explain certain principles of the invention and its practical application to enable one skilled in the art to make and use various exemplary embodiments of the invention and various alternatives and modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the claims and their equivalents.
Claims (9)
1. A dangerous behavior recognition method, comprising:
acquiring a real-time image of a target area;
inputting the real-time image into a dangerous behavior recognition model, and outputting a dangerous behavior recognition result of the real-time image;
the dangerous behavior recognition model is obtained by training according to a vehicle sample and a person sample and labels corresponding to the vehicle sample and the person sample based on a yolov5 algorithm.
2. The dangerous behavior recognition method according to claim 1, wherein after inputting the real-time image into a dangerous behavior recognition model and outputting a dangerous behavior recognition result of the real-time image, the method further comprises:
and sending an instruction corresponding to the target behavior to an execution mechanism under the condition that the dangerous behavior identification result is the target behavior, so that the execution mechanism executes the operation corresponding to the target behavior.
3. The dangerous behavior recognition method according to claim 1 or 2, wherein between the acquiring of the real-time image of the target area and the inputting of the real-time image into the dangerous behavior recognition model and the outputting of the dangerous behavior recognition result of the real-time image, the method further comprises:
and preprocessing the real-time image.
4. A dangerous behavior recognition apparatus, comprising:
the image acquisition module is used for acquiring a real-time image of the target area;
the target recognition module is used for inputting the real-time image into a dangerous behavior recognition model and outputting a dangerous behavior recognition result of the real-time image;
the dangerous behavior recognition model is obtained by training according to a vehicle sample and a person sample and labels corresponding to the vehicle sample and the person sample based on a yolov5 algorithm.
5. The hazardous behavior identification device of claim 4, further comprising:
and the instruction sending module is used for sending an instruction corresponding to the target behavior to an execution mechanism under the condition that the dangerous behavior identification result is the target behavior, so that the execution mechanism executes the operation corresponding to the target behavior.
6. The dangerous behavior recognition device according to claim 4 or 5, further comprising:
and the image preprocessing module is used for preprocessing the real-time image.
7. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the steps of the method for identifying dangerous behaviour according to any one of claims 1 to 3 are implemented when the program is executed by the processor.
8. A non-transitory computer-readable storage medium, on which a computer program is stored, wherein the computer program, when being executed by a processor, implements the steps of the hazardous behavior identification method according to any one of claims 1 to 3.
9. A computer program product comprising a computer program, wherein the computer program, when executed by a processor, performs the steps of the hazardous behavior identification method of any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210375655.1A CN114724246B (en) | 2022-04-11 | 2022-04-11 | Dangerous behavior identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210375655.1A CN114724246B (en) | 2022-04-11 | 2022-04-11 | Dangerous behavior identification method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114724246A true CN114724246A (en) | 2022-07-08 |
| CN114724246B CN114724246B (en) | 2024-01-30 |
Family
ID=82244182
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210375655.1A Active CN114724246B (en) | 2022-04-11 | 2022-04-11 | Dangerous behavior identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114724246B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115272701A (en) * | 2022-08-11 | 2022-11-01 | 电子科技大学 | Ship target detection method based on adaptive feature extraction and decoupling prediction head |
| CN115527045A (en) * | 2022-09-21 | 2022-12-27 | 北京拙河科技有限公司 | Image identification method and device for snow field danger identification |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101311947A (en) * | 2008-06-12 | 2008-11-26 | 浙江大学 | Real time intelligent control method based on natural video frequency |
| US20180268222A1 (en) * | 2017-03-17 | 2018-09-20 | Nec Laboratories America, Inc. | Action recognition system for action recognition in unlabeled videos with domain adversarial learning and knowledge distillation |
| CN110969130A (en) * | 2019-12-03 | 2020-04-07 | 厦门瑞为信息技术有限公司 | Driver dangerous action identification method and system based on YOLOV3 |
| CN112487972A (en) * | 2020-11-30 | 2021-03-12 | 福州外语外贸学院 | Anti-riot early warning method and device for office places in financial industry and storage medium |
| CN112818768A (en) * | 2021-01-19 | 2021-05-18 | 南京邮电大学 | Transformer substation reconstruction and extension violation behavior intelligent identification method based on meta-learning |
| CN112966589A (en) * | 2021-03-03 | 2021-06-15 | 中润油联天下网络科技有限公司 | Behavior identification method in dangerous area |
| CN113505709A (en) * | 2021-07-15 | 2021-10-15 | 开放智能机器（上海）有限公司 | Method and system for monitoring dangerous behaviors of human body in real time |
| CN113642474A (en) * | 2021-08-17 | 2021-11-12 | 四川航天电液控制有限公司 | Hazardous area personnel monitoring method based on YOLOV5 |
| CN113657211A (en) * | 2021-07-30 | 2021-11-16 | 浙江大华技术股份有限公司 | Early warning method, device and system for abnormal behavior and storage medium |
| WO2022001091A1 (en) * | 2020-06-29 | 2022-01-06 | 北京百度网讯科技有限公司 | Dangerous driving behavior recognition method and apparatus, and electronic device and storage medium |
| CN113901871A (en) * | 2021-09-07 | 2022-01-07 | 武汉未来幻影科技有限公司 | Driver dangerous action recognition method, device and equipment |
| CN114155601A (en) * | 2021-12-02 | 2022-03-08 | 山东中科先进技术研究院有限公司 | Vision-based method and system for detecting dangerous behaviors of operating personnel |
-
2022
- 2022-04-11 CN CN202210375655.1A patent/CN114724246B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101311947A (en) * | 2008-06-12 | 2008-11-26 | 浙江大学 | Real time intelligent control method based on natural video frequency |
| US20180268222A1 (en) * | 2017-03-17 | 2018-09-20 | Nec Laboratories America, Inc. | Action recognition system for action recognition in unlabeled videos with domain adversarial learning and knowledge distillation |
| CN110969130A (en) * | 2019-12-03 | 2020-04-07 | 厦门瑞为信息技术有限公司 | Driver dangerous action identification method and system based on YOLOV3 |
| WO2022001091A1 (en) * | 2020-06-29 | 2022-01-06 | 北京百度网讯科技有限公司 | Dangerous driving behavior recognition method and apparatus, and electronic device and storage medium |
| CN112487972A (en) * | 2020-11-30 | 2021-03-12 | 福州外语外贸学院 | Anti-riot early warning method and device for office places in financial industry and storage medium |
| CN112818768A (en) * | 2021-01-19 | 2021-05-18 | 南京邮电大学 | Transformer substation reconstruction and extension violation behavior intelligent identification method based on meta-learning |
| CN112966589A (en) * | 2021-03-03 | 2021-06-15 | 中润油联天下网络科技有限公司 | Behavior identification method in dangerous area |
| CN113505709A (en) * | 2021-07-15 | 2021-10-15 | 开放智能机器（上海）有限公司 | Method and system for monitoring dangerous behaviors of human body in real time |
| CN113657211A (en) * | 2021-07-30 | 2021-11-16 | 浙江大华技术股份有限公司 | Early warning method, device and system for abnormal behavior and storage medium |
| CN113642474A (en) * | 2021-08-17 | 2021-11-12 | 四川航天电液控制有限公司 | Hazardous area personnel monitoring method based on YOLOV5 |
| CN113901871A (en) * | 2021-09-07 | 2022-01-07 | 武汉未来幻影科技有限公司 | Driver dangerous action recognition method, device and equipment |
| CN114155601A (en) * | 2021-12-02 | 2022-03-08 | 山东中科先进技术研究院有限公司 | Vision-based method and system for detecting dangerous behaviors of operating personnel |
Non-Patent Citations (2)
| Title |
|---|
| 孟柯;吴超仲;陈志军;吕能超;邓超;刘钢;: "人车碰撞风险识别及智能车辆控制系统", 交通信息与安全, no. 06, pages 22 - 29 * |
| 张继奎等: "基于罪犯危险行为预测的视频报警机制的研究", 《中国新通信》, pages 108 - 109 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115272701A (en) * | 2022-08-11 | 2022-11-01 | 电子科技大学 | Ship target detection method based on adaptive feature extraction and decoupling prediction head |
| CN115272701B (en) * | 2022-08-11 | 2023-08-22 | 电子科技大学 | Ship target detection method based on self-adaptive feature extraction and decoupling prediction head |
| CN115527045A (en) * | 2022-09-21 | 2022-12-27 | 北京拙河科技有限公司 | Image identification method and device for snow field danger identification |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114724246B (en) | 2024-01-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112001902A (en) | Defect detection method and related device, equipment and storage medium | |
| CN111814902A (en) | Target detection model training method, target identification method, device and medium | |
| CN108806334A (en) | A kind of intelligent ship personal identification method based on image | |
| EP3690745B1 (en) | Learning method and learning device for providing functional safety by warning driver about potential dangerous situation by using explainable ai which verifies detection processes of autonomous driving network, and testing method and testing device using the same | |
| CN112036400B (en) | Method for constructing network for target detection and target detection method and system | |
| CN110033481A (en) | Method and apparatus for carrying out image procossing | |
| CN109389105B (en) | Multitask-based iris detection and visual angle classification method | |
| CN109815823B (en) | Data processing method and related product | |
| CN110059579B (en) | Method and apparatus for in vivo testing, electronic device, and storage medium | |
| CN114724246B (en) | Dangerous behavior identification method and device | |
| CN110991385A (en) | Method and device for identifying ship driving track and electronic equipment | |
| CN114926747A (en) | Remote sensing image directional target detection method based on multi-feature aggregation and interaction | |
| CN111709404B (en) | Machine room legacy identification method, system and equipment | |
| CN113569981A (en) | Power inspection bird nest detection method based on single-stage target detection network | |
| CN109167998A (en) | Detect method and device, the electronic equipment, storage medium of camera status | |
| CN115100614A (en) | Evaluation method and device of vehicle perception system, vehicle and storage medium | |
| CN111310751A (en) | License plate recognition method and device, electronic equipment and storage medium | |
| CN111127327B (en) | Picture inclination detection method and device | |
| CN111144361A (en) | Road lane detection method based on binaryzation CGAN network | |
| CN112084815A (en) | Target detection method based on camera focal length conversion, storage medium and processor | |
| CN113642353B (en) | Training method of face detection model, storage medium and terminal equipment | |
| CN115311680A (en) | Human body image quality detection method and device, electronic equipment and storage medium | |
| CN112927231B (en) | Training method of vehicle body dirt detection model, vehicle body dirt detection method and device | |
| Alfinnur Charisma | Modified YOLOv5-based License Plate Detection: an Efficient Approach for Automatic Vehicle Identification | |
| CN114743264B (en) | Shooting behavior detection method, device, equipment and storage medium |
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 |