CN113591963A - Equipment side loss detection model training method and equipment side loss detection method - Google Patents
Equipment side loss detection model training method and equipment side loss detection method Download PDFInfo
- Publication number
- CN113591963A CN113591963A CN202110837836.7A CN202110837836A CN113591963A CN 113591963 A CN113591963 A CN 113591963A CN 202110837836 A CN202110837836 A CN 202110837836A CN 113591963 A CN113591963 A CN 113591963A
- Authority
- CN
- China
- Prior art keywords
- equipment
- loss detection
- detection model
- equipment side
- loss
- 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 198
- 238000012549 training Methods 0.000 title claims abstract description 55
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000002372 labelling Methods 0.000 claims abstract description 37
- 238000013507 mapping Methods 0.000 claims abstract description 31
- 238000013528 artificial neural network Methods 0.000 claims abstract description 26
- 238000012360 testing method Methods 0.000 claims description 8
- 230000004913 activation Effects 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 230000002776 aggregation Effects 0.000 claims description 5
- 238000004220 aggregation Methods 0.000 claims description 5
- 238000005452 bending Methods 0.000 claims description 3
- 239000003973 paint Substances 0.000 claims description 3
- 230000006870 function Effects 0.000 description 17
- 238000003860 storage Methods 0.000 description 15
- 238000004590 computer program Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 9
- 238000011084 recovery Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000004931 aggregating effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/30—Administration of product recycling or disposal
-
- 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
- Y02W—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO WASTEWATER TREATMENT OR WASTE MANAGEMENT
- Y02W90/00—Enabling technologies or technologies with a potential or indirect contribution to greenhouse gas [GHG] emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Sustainable Development (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a training method and a detection method for a device side loss detection model. Further, a deep neural network is established according to the side image data set, and an equipment side loss detection model for detecting equipment side loss is obtained; and the detection result of the equipment side surface loss detection model has a mapping relation with the appearance type. Based on the method, different from the conventional training set labeling, the specific appearance type labeling is carried out on the appearance image of the side surface of the equipment, the loss type and the hardware type in the side surface of the equipment are accurately detected through the mapping relation, an equipment side surface loss detection model capable of being accurately detected is obtained, and the equipment side surface loss detection accuracy is improved while the equipment side surface loss detection workload is reduced.
Description
Technical Field
The invention relates to the technical field of electronic products, in particular to a training method of an equipment side loss detection model and an equipment side loss detection method.
Background
With the development of electronic product technology, various intelligent devices such as smart phones, notebook computers, tablet computers, and the like are developed. At present, along with the rapid development of economy and technology, the popularization and the updating speed of intelligent equipment are also faster and faster. Taking a smart phone as an example, the coming of the 5G era accelerates the generation change of the smart phone. In the iterative process of the intelligent equipment, effective recovery is one of effective utilization means of the residual value of the intelligent equipment, and the chemical pollution to the environment and the waste can be reduced.
In the recovery process of intelligent equipment, because the structure and the accessory of equipment side are comparatively complicated various, and the projected area of equipment side is little, brings very big degree of difficulty for the detection of equipment side. However, the loss of the side of the device often brings great adverse effects to the user experience, so the device side loss detection is one of the core detections of the recovery detection. At present, the traditional detection mode aiming at the side loss of the intelligent equipment mainly determines the accuracy degree of the side loss of the equipment through manual observation. However, the manual observation is time-consuming and labor-consuming, and it is difficult to ensure the accuracy and stability of the detection of the side loss of the device. Meanwhile, the accuracy of the side loss detection of the equipment is difficult to ensure on the premise that the projection area of the side of the equipment is small by using the existing modes of model detection, image recognition detection and the like.
In summary, it can be seen that the above disadvantages exist in the conventional detection method for the side loss of the smart device.
Disclosure of Invention
Therefore, it is necessary to provide an apparatus side loss detection model training method and an apparatus side loss detection method for overcoming the defects of the conventional detection method for the side loss of the smart apparatus.
A training method of an equipment side loss detection model comprises the following steps:
acquiring an equipment side surface appearance image of the intelligent equipment;
carrying out appearance type labeling on the side appearance image of the equipment to obtain a side image data set; wherein the appearance category comprises a wear category and a hardware category;
establishing a deep neural network according to the side image data set to obtain an equipment side loss detection model for detecting equipment side loss; and the detection result of the equipment side surface loss detection model has a mapping relation with the appearance type.
According to the equipment side loss detection model training method, after the equipment side appearance image of the intelligent equipment is obtained, appearance type labeling is carried out on the equipment side appearance image, and a side image data set is obtained. Further, a deep neural network is established according to the side image data set, and an equipment side loss detection model for detecting equipment side loss is obtained; and the detection result of the equipment side surface loss detection model has a mapping relation with the appearance type. Based on the method, different from the conventional training set labeling, the specific appearance type labeling is carried out on the appearance image of the side surface of the equipment, the loss type and the hardware type in the side surface of the equipment are accurately detected through the mapping relation, an equipment side surface loss detection model capable of being accurately detected is obtained, and the equipment side surface loss detection accuracy is improved while the equipment side surface loss detection workload is reduced.
In one embodiment, the deep neural network includes a cross-phase local network layer, a path aggregation network layer, and a final detection layer.
In one embodiment, the cross-phase local network layer comprises a backhaul network.
In one embodiment, the final detection layer includes an activation function, an optimization function, and a loss function.
In one embodiment, the side image data sets include a training set and a test set; wherein the ratio of the training set to the test set is 9: 1.
In one embodiment, the ratio of appearance categories is 1: 1.
In one embodiment, the wear category includes pop, paint drop, battery expansion, or body bending;
hardware classes include headphone jack, charging jack, volume button, SIM card slot, or stylus.
An apparatus side loss detection model training device, comprising:
the image acquisition module is used for acquiring an equipment side surface appearance image of the intelligent equipment;
the image labeling module is used for performing appearance type labeling on the side appearance image of the equipment to obtain a side image data set; wherein the appearance category comprises a wear category and a hardware category;
the model training module is used for establishing a deep neural network according to the side image data set to obtain an equipment side loss detection model for detecting the equipment side loss; and the detection result of the equipment side surface loss detection model has a mapping relation with the appearance type.
After the device side surface appearance image of the intelligent device is obtained, the device side surface appearance image is subjected to appearance type labeling to obtain a side surface image data set. Further, a deep neural network is established according to the side image data set, and an equipment side loss detection model for detecting equipment side loss is obtained; and the detection result of the equipment side surface loss detection model has a mapping relation with the appearance type. Based on the method, different from the conventional training set labeling, the specific appearance type labeling is carried out on the appearance image of the side surface of the equipment, the loss type and the hardware type in the side surface of the equipment are accurately detected through the mapping relation, an equipment side surface loss detection model capable of being accurately detected is obtained, and the equipment side surface loss detection accuracy is improved while the equipment side surface loss detection workload is reduced.
A computer storage medium having computer instructions stored thereon, the computer instructions when executed by a processor implementing the device side loss detection model training method of any of the above embodiments.
After the computer storage medium obtains the device side appearance image of the intelligent device, appearance type labeling is carried out on the device side appearance image to obtain a side image data set. Further, a deep neural network is established according to the side image data set, and an equipment side loss detection model for detecting equipment side loss is obtained; and the detection result of the equipment side surface loss detection model has a mapping relation with the appearance type. Based on the method, different from the conventional training set labeling, the specific appearance type labeling is carried out on the appearance image of the side surface of the equipment, the loss type and the hardware type in the side surface of the equipment are accurately detected through the mapping relation, an equipment side surface loss detection model capable of being accurately detected is obtained, and the equipment side surface loss detection accuracy is improved while the equipment side surface loss detection workload is reduced.
A computer device includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, the method for training a device side wear detection model according to any of the embodiments described above is implemented.
After the computer device obtains the device side surface appearance image of the intelligent device, appearance type labeling is carried out on the device side surface appearance image to obtain a side surface image data set. Further, a deep neural network is established according to the side image data set, and an equipment side loss detection model for detecting equipment side loss is obtained; and the detection result of the equipment side surface loss detection model has a mapping relation with the appearance type. Based on the method, different from the conventional training set labeling, the specific appearance type labeling is carried out on the appearance image of the side surface of the equipment, the loss type and the hardware type in the side surface of the equipment are accurately detected through the mapping relation, an equipment side surface loss detection model capable of being accurately detected is obtained, and the equipment side surface loss detection accuracy is improved while the equipment side surface loss detection workload is reduced.
A device side loss detection method comprises the following steps:
acquiring an equipment side image of the intelligent equipment to be tested;
and inputting the device side image into the device side loss detection model to obtain a detection result for representing the device side loss of the intelligent device to be detected.
According to the equipment side loss detection method, after the equipment side image of the intelligent equipment to be detected is obtained, the equipment side image is input into the equipment side loss detection model, and a detection result used for representing the equipment side loss of the intelligent equipment to be detected is obtained. Based on the method, the loss category and the hardware category in the side face of the equipment are accurately detected through the mapping relation of the equipment side face loss detection model, and the accuracy of equipment side face loss detection is improved while the workload of equipment side face loss detection is reduced.
An apparatus side loss detection device, comprising:
the image acquisition module is used for acquiring an equipment side image of the intelligent equipment to be tested;
and the loss detection module is used for inputting the device side image into the device side loss detection model to obtain a detection result for representing the device side loss of the intelligent device to be detected.
After the device side image of the intelligent device to be detected is obtained, the device side image is input into the device side loss detection model, and a detection result for representing the device side loss of the intelligent device to be detected is obtained. Based on the method, the loss category and the hardware category in the side face of the equipment are accurately detected through the mapping relation of the equipment side face loss detection model, and the accuracy of equipment side face loss detection is improved while the workload of equipment side face loss detection is reduced.
A computer storage medium having computer instructions stored thereon, the computer instructions when executed by a processor implementing the device side wear detection method of any of the above embodiments.
After the computer storage medium obtains the device side image of the intelligent device to be detected, the device side image is input into the device side loss detection model, and a detection result used for representing the device side loss of the intelligent device to be detected is obtained. Based on the method, the loss category and the hardware category in the side face of the equipment are accurately detected through the mapping relation of the equipment side face loss detection model, and the accuracy of equipment side face loss detection is improved while the workload of equipment side face loss detection is reduced.
A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the device side wear detection method of any of the above embodiments when executing the program.
After the computer device obtains the device side image of the intelligent device to be detected, the device side image is input into the device side loss detection model, and a detection result used for representing the device side loss of the intelligent device to be detected is obtained. Based on the method, the loss category and the hardware category in the side face of the equipment are accurately detected through the mapping relation of the equipment side face loss detection model, and the accuracy of equipment side face loss detection is improved while the workload of equipment side face loss detection is reduced.
Drawings
FIG. 1 is a flow chart of an apparatus side loss detection model training method according to an embodiment;
FIG. 2 is a flow chart of an apparatus side loss detection model training method according to another embodiment;
FIG. 3 is a schematic diagram of a deep neural network architecture according to an embodiment;
FIG. 4 is a block diagram of an apparatus side loss detection model training device according to an embodiment;
FIG. 5 is a flow diagram of a method for detecting device side loss according to one embodiment;
FIG. 6 is a block diagram of an apparatus side loss detection device according to an embodiment;
FIG. 7 is a schematic diagram of an internal structure of a computer according to an embodiment.
Detailed Description
For better understanding of the objects, technical solutions and effects of the present invention, the present invention will be further explained with reference to the accompanying drawings and examples. Meanwhile, the following described examples are only for explaining the present invention, and are not intended to limit the present invention.
The embodiment of the invention provides a training method of an equipment side loss detection model.
Fig. 1 is a flowchart of an apparatus side loss detection model training method according to an embodiment, and as shown in fig. 1, the apparatus side loss detection model training method according to an embodiment includes steps S100 to S102:
s100, acquiring an equipment side surface appearance image of the intelligent equipment;
s101, performing appearance type labeling on the side appearance image of the equipment to obtain a side image data set; wherein the appearance category comprises a wear category and a hardware category;
s102, establishing a deep neural network according to the side image data set to obtain an equipment side loss detection model for detecting equipment side loss; and the detection result of the equipment side surface loss detection model has a mapping relation with the appearance type.
The intelligent device comprises multiple types of devices, such as an ios device, an android device or a mobile computer device. The intelligent equipment is used as a training sample source of the equipment side loss detection model, and the equipment side appearance images of various types of intelligent equipment can be obtained as much as possible to serve as historical data, so that the accuracy of the subsequent equipment side loss detection model is improved.
In one embodiment, in the recovery process of the intelligent device, the side face of the intelligent device can be shot by the shooting device through a self-service terminal or a recovery machine, and the side face appearance image of the intelligent device is collected. And acquiring an appearance image of the side surface of the equipment after the shooting of the shooting equipment is finished. The self-service terminal or the recovery machine serves as an execution main body to complete the execution of the equipment side loss detection model training method of one embodiment, or the cloud server transmits the equipment side appearance image to the cloud server and serves as the execution main body to complete the execution of the equipment side loss detection model training method of one embodiment.
And further, performing appearance category labeling on the side appearance image of the equipment, and establishing a side image data set comprising a training set and a testing set. The method of the embodiment is different from the traditional image labeling process of image detection or model detection, and performs one-to-one type of appearance type labeling on the side appearance of the intelligent device, so that more appearance types are determined when more side appearance images are generated on the device, and the mapping relation of a subsequent device side loss detection model is enriched.
The appearance category comprises a loss category and a hardware category, the loss category comprises loss types or loss accessories of the intelligent device, and the hardware category comprises power distribution types on the side face of the intelligent device.
In one embodiment, taking a smart device as a mobile phone as an example, the loss category includes explosion, paint dropping, battery expansion or body bending;
hardware classes include headphone jack, charging jack, volume button, SIM card slot, or stylus.
And marking the appearance type of each position of the side appearance image of the equipment through appearance type marking to obtain a side image data set.
In one embodiment, fig. 2 is a flowchart of a training method for a device side loss detection model according to another embodiment, and as shown in fig. 2, a process of performing appearance class labeling on a device side appearance image in step S101 to obtain a side image data set includes step S200:
and S200, performing appearance type labeling on the side appearance image of the equipment through third-party open source software.
In one embodiment, the sensitivity of the device side loss detection model to appearance categories is adjusted by adjusting the scale of the appearance categories. Wherein, the proportion of each appearance category is higher according to the sensitivity requirement, and the proportion of the more sensitive appearance category is higher. As a preferred embodiment, the ratio of each appearance type is 1: 1.
Meanwhile, in the side image data set, the proportion of the training set and the test set is correspondingly divided according to the requirements of modeling training. In one embodiment, the duty cycle of the training set is higher than the duty cycle of the test set. As a preferred embodiment, the ratio of training set to test set is 9: 1.
Based on the method, a deep neural network is established according to the side image data set, and the training requirement of the side appearance image of the equipment with high data volume and the establishment of the mapping relation are met. In one embodiment, the deep neural network comprises a YOLOv5 deep neural network or a YOLOv3 deep neural network. As a preferred embodiment, the deep neural network is the YOLOv5 deep neural network used under the framework of a pyrrch.
In one embodiment, the deep neural network includes a cross-phase local network layer, a path aggregation network layer, and a final detection layer.
Fig. 3 is a schematic diagram of a deep neural network structure layer according to an embodiment, and as shown in fig. 3, after an apparatus side image of an intelligent apparatus to be tested is input, a cross-stage local network layer integrates a gradient change into a feature map from beginning to end to reduce the parameter number of a model and a flo (computational power) value. Therefore, the inference speed and the accuracy are ensured, and the size of a side loss detection model of subsequent equipment is reduced.
In one embodiment, the cross-phase local network layer includes a backhaul network, such as csprassenext 50 and CSPDarknet 53. As a preferred embodiment, the cross-phase local network layer uses CSPDarknet 53.
The path aggregation network layer is used for enhancing information propagation, recovering damaged information paths between each candidate frame region and all the feature layers by utilizing self-adaptive feature pooling, aggregating each candidate region on each feature layer, and avoiding being randomly distributed so as to enhance the device side loss detection model.
In one embodiment, the path aggregation network layer aggregates features using PANET as the Neck.
In one embodiment, the final detection layer includes an activation function, an optimization function, and a loss function.
The activation function is used for calculating the probability of each appearance category, and the optimal parameter value is obtained by the optimization function, and the model accuracy is calculated by the loss function.
As a preferred embodiment, the middle hidden layer uses a Relu activation function, and the final detection layer uses a Sigmoid activation function; the optimization function comprises an Adam function and an SGD function; the loss functions include binary cross entropy and logs loss functions.
An anchor box is applied to the feature map through the layers of the deep neural network and a final output vector with class probabilities, object scores and bounding boxes is generated. And then, calculating by using an optimization function and a loss function to obtain an optimal weight parameter, and obtaining an equipment side loss detection model. Based on this, the correspondence between the detection result and the appearance type in the mapping relationship is realized.
In the method for training the equipment side loss detection model in any embodiment, after the equipment side appearance image of the intelligent equipment is obtained, appearance category labeling is performed on the equipment side appearance image, so that a side image data set is obtained. Further, a deep neural network is established according to the side image data set, and an equipment side loss detection model for detecting equipment side loss is obtained; and the detection result of the equipment side surface loss detection model has a mapping relation with the appearance type. Based on the method, different from the conventional training set labeling, the specific appearance type labeling is carried out on the appearance image of the side surface of the equipment, the loss type and the hardware type in the side surface of the equipment are accurately detected through the mapping relation, an equipment side surface loss detection model capable of being accurately detected is obtained, and the equipment side surface loss detection accuracy is improved while the equipment side surface loss detection workload is reduced.
The embodiment of the invention provides a training device for a side loss detection model of equipment.
Fig. 4 is a block diagram of an apparatus side loss detection model training apparatus according to an embodiment, and as shown in fig. 4, the apparatus side loss detection model training apparatus according to an embodiment includes a module 100, a module 101, and a module 102:
the image acquisition module 100 is used for acquiring an equipment side appearance image of the intelligent equipment;
the image labeling module 101 is configured to perform appearance category labeling on an appearance image of a side of the device to obtain a side image dataset; wherein the appearance category comprises a wear category and a hardware category;
the model training module 102 is used for establishing a deep neural network according to the side image data set to obtain an equipment side loss detection model for detecting the equipment side loss; and the detection result of the equipment side surface loss detection model has a mapping relation with the appearance type.
After the device side surface appearance image of the intelligent device is obtained, the device side surface appearance image is subjected to appearance type labeling to obtain a side surface image data set. Further, a deep neural network is established according to the side image data set, and an equipment side loss detection model for detecting equipment side loss is obtained; and the detection result of the equipment side surface loss detection model has a mapping relation with the appearance type. Based on the method, different from the conventional training set labeling, the specific appearance type labeling is carried out on the appearance image of the side surface of the equipment, the loss type and the hardware type in the side surface of the equipment are accurately detected through the mapping relation, an equipment side surface loss detection model capable of being accurately detected is obtained, and the equipment side surface loss detection accuracy is improved while the equipment side surface loss detection workload is reduced.
The embodiment of the invention also provides a device side loss detection method.
Fig. 5 is a flowchart of an apparatus side loss detection method according to an embodiment, and as shown in fig. 5, the apparatus side loss detection method according to an embodiment includes steps S300 and S301:
s300, acquiring an equipment side image of the intelligent equipment to be tested;
and S301, inputting the device side image into the device side loss detection model to obtain a detection result for representing the device side loss of the intelligent device to be detected.
After the equipment side loss detection model is determined in the steps, the equipment side image is input as a feature map through an open interface of the equipment side loss detection model, operation is executed in the equipment side loss detection model, and a detection result is obtained and transmitted to a corresponding rear-end interface so as to complete side loss detection of the intelligent equipment to be detected.
According to the equipment side loss detection method, after the equipment side image of the intelligent equipment to be detected is obtained, the equipment side image is input into the equipment side loss detection model, and a detection result used for representing the equipment side loss of the intelligent equipment to be detected is obtained. Based on the method, the loss category and the hardware category in the side face of the equipment are accurately detected through the mapping relation of the equipment side face loss detection model, and the accuracy of equipment side face loss detection is improved while the workload of equipment side face loss detection is reduced.
The embodiment of the invention also provides a device for detecting the side loss of the equipment.
Fig. 6 is a block diagram of the device side loss detection apparatus according to the embodiment, and as shown in fig. 6, the device side loss detection apparatus according to the embodiment includes a block 200 and a block 201:
the image acquisition module 200 is used for acquiring an equipment side image of the intelligent equipment to be tested;
and the loss detection module 201 is configured to input the device side image into the device side loss detection model, and obtain a detection result for representing the device side loss of the to-be-detected intelligent device.
After the device side image of the intelligent device to be detected is obtained, the device side image is input into the device side loss detection model, and a detection result for representing the device side loss of the intelligent device to be detected is obtained. Based on the method, the loss category and the hardware category in the side face of the equipment are accurately detected through the mapping relation of the equipment side face loss detection model, and the accuracy of equipment side face loss detection is improved while the workload of equipment side face loss detection is reduced.
The embodiment of the present invention further provides a computer storage medium, on which computer instructions are stored, and when the instructions are executed by a processor, the method for training a device side loss detection model or the method for detecting a device side loss according to any of the above embodiments is implemented.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware related to instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, the computer program can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
Alternatively, the integrated unit of the present invention may be stored in a computer-readable storage medium if it is implemented in the form of a software functional module and sold or used as a separate product. Based on such understanding, the technical solutions of the embodiments 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 terminal, or a network device) to execute all or part of the methods of the embodiments of the present invention. And the aforementioned storage medium includes: a removable storage device, a RAM, a ROM, a magnetic or optical disk, or various other media that can store program code.
Corresponding to the computer storage medium, in an embodiment, there is also provided a computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any one of the device side wear detection model training method and the device side wear detection method in the embodiments.
The computer device may be a terminal, and its internal structure diagram may be as shown in fig. 7. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a device side loss detection model training method or a device side loss detection method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
After the computer device obtains the device side surface appearance image of the intelligent device, appearance type labeling is carried out on the device side surface appearance image to obtain a side surface image data set. Further, a deep neural network is established according to the side image data set, and an equipment side loss detection model for detecting equipment side loss is obtained; and the detection result of the equipment side surface loss detection model has a mapping relation with the appearance type. Based on the method, different from the conventional training set labeling, the specific appearance type labeling is carried out on the appearance image of the side surface of the equipment, the loss type and the hardware type in the side surface of the equipment are accurately detected through the mapping relation, an equipment side surface loss detection model capable of being accurately detected is obtained, and the equipment side surface loss detection accuracy is improved while the equipment side surface loss detection workload is reduced.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above examples only show some embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A training method for an equipment side loss detection model is characterized by comprising the following steps:
acquiring an equipment side surface appearance image of the intelligent equipment;
performing appearance category labeling on the side appearance image of the equipment to obtain a side image data set; wherein the appearance categories include a wear category and a hardware category;
establishing a deep neural network according to the side image data set to obtain an equipment side loss detection model for detecting equipment side loss; and the detection result of the equipment side loss detection model has a mapping relation with the appearance type.
2. The device side loss detection model training method of claim 1, wherein the deep neural network comprises a cross-stage local network layer, a path aggregation network layer, and a final detection layer.
3. The device side loss detection model training method of claim 2, wherein the cross-phase local network layer comprises a backhaul network.
4. The device side loss detection model training method of claim 2, wherein the final detection layer comprises an activation function, an optimization function, and a loss function.
5. The device side loss detection model training method according to any one of claims 1 to 4, wherein the side image data set includes a training set and a test set; wherein the ratio of the training set to the test set is 9: 1.
6. The device side loss detection model training method according to any one of claims 1 to 4, wherein the ratio of each appearance category is 1: 1.
7. The training method for the equipment side loss detection model according to any one of claims 1 to 4, wherein the loss category comprises outbreak, paint drop, battery expansion or body bending;
the hardware category includes an earphone hole, a charging hole, a volume button, a SIM card slot, or a stylus.
8. A method for detecting the side loss of equipment is characterized by comprising the following steps:
acquiring an equipment side image of the intelligent equipment to be tested;
and inputting the equipment side image into an equipment side loss detection model to obtain a detection result for representing the equipment side loss of the intelligent equipment to be detected.
9. The utility model provides an equipment side loss detection model trainer which characterized in that includes:
the image acquisition module is used for acquiring an equipment side surface appearance image of the intelligent equipment;
the image labeling module is used for performing appearance type labeling on the side appearance image of the equipment to obtain a side image data set; wherein the appearance categories include a wear category and a hardware category;
the model training module is used for establishing a deep neural network according to the side image data set to obtain an equipment side loss detection model for detecting equipment side loss; and the detection result of the equipment side loss detection model has a mapping relation with the appearance type.
10. An apparatus side loss detection device, comprising:
the image acquisition module is used for acquiring an equipment side image of the intelligent equipment to be tested;
and the loss detection module is used for inputting the equipment side image into the equipment side loss detection model to obtain a detection result for representing the equipment side loss of the intelligent equipment to be detected.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110837836.7A CN113591963A (en) | 2021-07-23 | 2021-07-23 | Equipment side loss detection model training method and equipment side loss detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110837836.7A CN113591963A (en) | 2021-07-23 | 2021-07-23 | Equipment side loss detection model training method and equipment side loss detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113591963A true CN113591963A (en) | 2021-11-02 |
Family
ID=78249306
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110837836.7A Pending CN113591963A (en) | 2021-07-23 | 2021-07-23 | Equipment side loss detection model training method and equipment side loss detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113591963A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11922467B2 (en) | 2020-08-17 | 2024-03-05 | ecoATM, Inc. | Evaluating an electronic device using optical character recognition |
| US12033454B2 (en) | 2020-08-17 | 2024-07-09 | Ecoatm, Llc | Kiosk for evaluating and purchasing used electronic devices |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111294454A (en) * | 2020-03-23 | 2020-06-16 | 广州绿怡信息科技有限公司 | Mobile phone loss detection device and method |
| CN111898651A (en) * | 2020-07-10 | 2020-11-06 | 江苏科技大学 | Tree detection method based on Tiny Yolov3 algorithm |
| CN112288074A (en) * | 2020-08-07 | 2021-01-29 | 京东安联财产保险有限公司 | Image recognition network generation method and device, storage medium and electronic equipment |
| CN112614117A (en) * | 2020-12-28 | 2021-04-06 | 广州绿怡信息科技有限公司 | Equipment region extraction model training method, equipment region extraction method and device |
| CN112634245A (en) * | 2020-12-28 | 2021-04-09 | 广州绿怡信息科技有限公司 | Loss detection model training method, loss detection method and device |
-
2021
- 2021-07-23 CN CN202110837836.7A patent/CN113591963A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111294454A (en) * | 2020-03-23 | 2020-06-16 | 广州绿怡信息科技有限公司 | Mobile phone loss detection device and method |
| CN111898651A (en) * | 2020-07-10 | 2020-11-06 | 江苏科技大学 | Tree detection method based on Tiny Yolov3 algorithm |
| CN112288074A (en) * | 2020-08-07 | 2021-01-29 | 京东安联财产保险有限公司 | Image recognition network generation method and device, storage medium and electronic equipment |
| CN112614117A (en) * | 2020-12-28 | 2021-04-06 | 广州绿怡信息科技有限公司 | Equipment region extraction model training method, equipment region extraction method and device |
| CN112634245A (en) * | 2020-12-28 | 2021-04-09 | 广州绿怡信息科技有限公司 | Loss detection model training method, loss detection method and device |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11922467B2 (en) | 2020-08-17 | 2024-03-05 | ecoATM, Inc. | Evaluating an electronic device using optical character recognition |
| US12033454B2 (en) | 2020-08-17 | 2024-07-09 | Ecoatm, Llc | Kiosk for evaluating and purchasing used electronic devices |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102635987B1 (en) | Method, apparatus, device and storage medium for training an image semantic segmentation network | |
| US11538246B2 (en) | Method and apparatus for training feature extraction model, computer device, and computer-readable storage medium | |
| CN111476306B (en) | Object detection method, device, equipment and storage medium based on artificial intelligence | |
| EP3944145A2 (en) | Method and device for training image recognition model, equipment and medium | |
| CN114155244B (en) | Defect detection method, device, equipment and storage medium | |
| CN110781970B (en) | Classifier generation method, device, equipment and storage medium | |
| CN112508092A (en) | Sample screening method, system, equipment and medium | |
| CN113298078A (en) | Equipment screen fragmentation detection model training method and equipment screen fragmentation detection method | |
| CN113591963A (en) | Equipment side loss detection model training method and equipment side loss detection method | |
| CN110309297A (en) | Rubbish text detection method, readable storage medium storing program for executing and computer equipment | |
| WO2024083121A1 (en) | Data processing method and apparatus | |
| CN113705297A (en) | Training method and device for detection model, computer equipment and storage medium | |
| CN113132523B (en) | Call detection model training method and call detection method | |
| CN113065634B (en) | Image processing method, neural network training method and related equipment | |
| CN112308069A (en) | Click test method, device, equipment and storage medium for software interface | |
| CN112634245A (en) | Loss detection model training method, loss detection method and device | |
| US20230385597A1 (en) | Multi-granularity perception integrated learning method, device, computer equipment and medium | |
| CN111444802A (en) | Face recognition method and device and intelligent terminal | |
| EP3913533A2 (en) | Method and apparatus of processing image device and medium | |
| CN114298345A (en) | Recovery valuation output model training method and recovery valuation output method | |
| CN114330533A (en) | Equipment screen aging two-classification model training method and equipment screen aging detection method | |
| CN111738290B (en) | Image detection method, model construction and training method, device, equipment and medium | |
| CN113052798A (en) | Screen aging detection model training method and screen aging detection method | |
| WO2022016650A1 (en) | Smart pen image processing method, apparatus and electronic device | |
| CN113237473A (en) | Compass function detection method and device |
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 |