Method for detecting recognition efficiency of electric police equipment and electronic equipment

Abstract

The present disclosure provides a method and electronic device for detecting the recognition efficiency of electric police equipment. It is used to improve the detection efficiency of electric police equipment, including: for any electric police equipment, periodically obtaining the target video shot by the electric police equipment in this period; for any image in the target video, using the target detection neural network model to perform vehicle detection on the image to obtain the position of each vehicle; deleting the target vehicle from the image to obtain the target image, wherein the target vehicle is determined by the position of each vehicle in the image and the position of each vehicle in the next frame image, and there are duplicate vehicles with each vehicle in the next frame image of the image; according to the total number of each vehicle in each target image, the actual number of each vehicle in this period is obtained; through the actual number and the detection number, it is determined whether the recognition efficiency of the electric police equipment is reduced, wherein the detection number is the number of vehicles recognized by the electric police equipment in this period.

Description

Background Art

[0002] Since general electric police equipment will have inaccurate target capture after being used for a certain number of years, affecting normal target recognition and capture as well as video monitoring functions, it is necessary to test the recognition efficiency of the electric police equipment to determine whether the electric police equipment needs to be replaced.

[0003] In the prior art, the recognition efficiency of electric police equipment is detected by manual verification, which results in low detection efficiency of electric police equipment. Summary of the invention

[0004] In an exemplary embodiment of the present disclosure, a method for detecting the recognition efficiency of electric police equipment and an electronic device are provided to improve the detection efficiency of electric police equipment.

[0005] A first aspect of the present disclosure provides an electronic device, including a processor and a storage unit, wherein:

[0006] The storage unit is configured to store the video captured by the electronic police equipment;

[0007] The processor is configured to:

[0008] For any electric police device, periodically obtain the target video shot by the electric police device in this period;

[0009] For any frame image in the target video, use the target detection neural network model to perform vehicle detection on the target detection area in the image to obtain the position of each vehicle;

[0010] Deleting a target vehicle from the image to obtain a target image, wherein the target vehicle is determined by the position of each vehicle in the image and the position of each vehicle in a next frame of the image, and there is a duplicate vehicle with each vehicle in the next frame of the image;

[0011] According to the total number of each vehicle in each target image, the actual number of each vehicle in this cycle is obtained;

[0012] Whether the recognition efficiency of the electric police equipment is reduced is determined by the actual number and the detection number, wherein the detection number is the number of vehicles recognized by the electric police equipment in this cycle.

[0013] In this embodiment, the target detection neural network model is used to detect the position of each vehicle in each image taken by the electric police device, and then the vehicles that are repeated in the next frame of each image are deleted from each image to obtain the actual number of each vehicle, and the actual number is compared with the detected number to determine whether the recognition efficiency of the electric police device is reduced. Therefore, in this embodiment, the recognition efficiency of the electric police device is automatically identified to see whether it is reduced, and no manual verification is required, thereby improving the detection efficiency of the electric police device.

[0014] In one embodiment, the target detection neural network model uses the content security policy dark network CSPdarknet53 as the backbone network, and the convolution layer of at least one residual block in the fourth residual block and the fifth residual block in the CSPdarknet53 adopts depth-separable convolution.

[0015] In this embodiment, the convolutional layer of at least one residual block in the fourth residual block and the fifth residual block in the backbone network CSPdarknet53 is replaced by a depthwise separable convolution. Since the depthwise separable convolution requires less computation than the standard convolution, the detection efficiency of the electric police equipment can be further improved.

[0016] In one embodiment, the activation function in the depthwise separable convolution is a mish activation function.

[0017] In this embodiment, the original activation function in the depthwise separable convolution is replaced with the mish activation function because the mish activation function is a smooth non-monotonic activation function with no upper bound and a lower bound, which can avoid the gradient saturation phenomenon and improve the regularization effect of the target detection neural network model. At the same time, the mish activation function allows a relatively small negative gradient to flow in, ensuring the flow of information, thereby improving the accuracy of the detection of electric police equipment.

[0018] In one embodiment, the processor is further configured to:

[0019] The target vehicle in the image is determined by:

[0020] For any vehicle in the image, perform the following steps:

[0021] According to the position of the arbitrary vehicle and the position of each vehicle in the next frame of the image, obtaining each distance between the arbitrary vehicle and each vehicle in the next frame of the image;

[0022] If there is a distance among the distances that satisfies a preset distance range, it is determined that any one of the vehicles is a target vehicle in the image.

[0023] In this embodiment, the position of any vehicle in an image and the positions of each vehicle in the next frame of the image are combined to obtain the distances between the any vehicle and each vehicle in the next frame of the image, and then the target vehicle in the image is determined by the distances between the vehicles. Thus, the accuracy of identifying the target vehicle is improved.

[0024] In one embodiment, the processor performs the step of determining whether the recognition efficiency of the electric police device is reduced by using the actual number and the detected number, and is specifically configured as follows:

[0025] Determining the recognition efficiency of the electric police device according to the actual number and the detected number;

[0026] Comparing the recognition efficiency with a preset standard recognition efficiency;

[0027] If the recognition efficiency is less than the standard recognition efficiency, it is determined that the recognition efficiency of the electric police device is reduced;

[0028] If the recognition efficiency is not less than the standard recognition efficiency, it is determined that the recognition efficiency of the electronic police equipment has not decreased.

[0029] In this embodiment, the recognition efficiency of the electric police device is determined by the actual number and the detected number, and the recognition efficiency is compared with the standard recognition efficiency to determine whether the recognition efficiency of the electric police device is reduced, thereby improving the accuracy of the recognition result.

[0030] A second aspect of the present disclosure provides a method for detecting recognition efficiency of an electric police device, the method comprising:

[0031] For any electric police device, periodically obtain the target video shot by the electric police device in this period;

[0032] For any frame image in the target video, use the target detection neural network model to perform vehicle detection on the target detection area in the image to obtain the position of each vehicle;

[0033] Deleting a target vehicle from the image to obtain a target image, wherein the target vehicle is determined by the position of each vehicle in the image and the position of each vehicle in a next frame of the image, and there is a duplicate vehicle with each vehicle in the next frame of the image;

[0034] According to the total number of each vehicle in each target image, the actual number of each vehicle in this cycle is obtained;

[0035] Whether the recognition efficiency of the electric police equipment is reduced is determined by the actual number and the detection number, wherein the detection number is the number of vehicles recognized by the electric police equipment in this cycle.

[0036] In one embodiment, the target detection neural network model uses the content security policy dark network CSPdarknet53 as the backbone network, and the convolution layer of at least one residual block in the fourth residual block and the fifth residual block in the CSPdarknet53 adopts depth-separable convolution.

[0037] In one embodiment, the activation function in the depthwise separable convolution is a mish activation function.

[0038] In one embodiment, the target vehicle in the image is determined by:

[0039] For any vehicle in the image, perform the following steps:

[0040] According to the position of the arbitrary vehicle and the position of each vehicle in the next frame of the image, obtaining each distance between the arbitrary vehicle and each vehicle in the next frame of the image;

[0041] If there is a distance among the distances that satisfies a preset distance range, it is determined that any one of the vehicles is a target vehicle in the image.

[0042] In one embodiment, determining whether the recognition efficiency of the electric police device is reduced by using the actual number and the detected number includes:

[0043] Determining the recognition efficiency of the electric police device according to the actual number and the detected number;

[0044] Comparing the recognition efficiency with a preset standard recognition efficiency;

[0045] If the recognition efficiency is less than the standard recognition efficiency, it is determined that the recognition efficiency of the electric police device is reduced;

[0046] If the recognition efficiency is not less than the standard recognition efficiency, it is determined that the recognition efficiency of the electronic police equipment has not decreased.

[0047] According to a third aspect provided by an embodiment of the present disclosure, a computer storage medium is provided, wherein the computer storage medium stores a computer program, and the computer program is used to execute the method described in the second aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0049] Figure 1 A schematic diagram of an applicable scenario in an embodiment of the present disclosure;

[0050] Figure 2 This is one of the flow charts of a method for detecting the recognition efficiency of an electric police device according to an embodiment of the present disclosure;

[0051] Figure 3 It is a schematic diagram of the structure of CSPdarknet53 according to one embodiment of the present disclosure;

[0052] Figure 4 Schematic diagram of the structure of depth-wise separable convolution according to an embodiment of the present disclosure;

[0053] Figure 5 A schematic diagram of the structure of a target detection neural network model YOLOv4 according to an embodiment of the present disclosure;

[0054] Figure 6 is a schematic diagram of the position of a vehicle in each frame image according to an embodiment of the present disclosure;

[0055] Figure 7 The second flowchart of the method for detecting the recognition efficiency of electric police equipment according to an embodiment of the present disclosure;

[0056] Figure 8 A device for detecting recognition efficiency of electric police equipment according to an embodiment of the present disclosure;

[0057] Fig. 9 The figure is a schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0059] In the embodiments of the present disclosure, the term "and / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent three situations: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship.

[0060] The application scenarios described in the embodiments of the present disclosure are intended to more clearly illustrate the technical solutions of the embodiments of the present disclosure, and do not constitute a limitation on the technical solutions provided by the embodiments of the present disclosure. It is known to those skilled in the art that with the emergence of new application scenarios, the technical solutions provided by the embodiments of the present disclosure are also applicable to similar technical problems. In the description of the present disclosure, unless otherwise specified, the meaning of "multiple" is two or more.

[0061] In the prior art, the recognition efficiency of electric police equipment is detected by manual verification, which results in low detection efficiency of electric police equipment.

[0062] Therefore, the present disclosure provides a method for detecting the recognition efficiency of electric police equipment, by using a target detection neural network model to detect the position of each vehicle in each image taken by the electric police equipment, and then deleting from each image the vehicles that are duplicated with each vehicle in the next frame of each image, thereby obtaining the actual number of each vehicle, and comparing the actual number with the detected number to determine whether the recognition efficiency of the electric police equipment is reduced. Therefore, in this embodiment, whether the recognition efficiency of the electric police equipment is reduced is automatically identified, and no manual verification is required, thereby improving the detection efficiency of the electric police equipment. Below, the scheme of the present disclosure is described in detail in conjunction with the accompanying drawings.

[0063] like Figure 1 As shown, an application scenario of a method for detecting the recognition efficiency of an electric police device is described, in which the electronic device is used as a server as an example. The application scenario includes multiple electric police devices 110, terminal devices 120 and servers 130. Figure 1 In the figure, two electric police devices 110 and one terminal device 120 are used as an example. In fact, the number of electric police devices 110 and terminal devices 120 is not limited. The terminal device 120 can be a mobile phone, a tablet computer, a personal computer, etc. The server 130 can be implemented by a single server or multiple servers. The server 130 can be implemented by a physical server or a virtual server.

[0064] In a possible application scenario, the server 130 periodically obtains the target video shot by the electric police device 110 in this cycle for any one of the electric police devices 110, and then the server 130 uses the target detection neural network model to detect vehicles in the target detection area in the image for any frame of the target video to obtain the position of each vehicle; and deletes the target vehicle from the image to obtain a target image, wherein the target vehicle is determined by the position of each vehicle in the image and the position of each vehicle in the next frame of the image, and there are duplicate vehicles with each vehicle in the next frame of the image. The server 130 obtains the actual number of each vehicle in this cycle based on the total number of each vehicle in each target image, and then determines whether the recognition efficiency of the electric police device is reduced based on the actual number and the number of vehicles recognized by the electric police device in this cycle, and sends the result of whether the recognition efficiency of the electric police device is reduced to the terminal device 120 for display.

[0065] like Figure 2 FIG. 1 is a flow chart of a method for detecting the recognition efficiency of an electric police device disclosed in the present invention, which may include the following steps:

[0066] Step 201: for any electric police device, periodically obtain a target video shot by the electric police device in this period;

[0067] Step 202: for any frame of the target video, use the target detection neural network model to perform vehicle detection on the target detection area in the image to obtain the position of each vehicle;

[0068] Among them, the target detection neural network model in this embodiment includes but is not limited to R-CNN (Region-Convolutional Neural Networks, regional convolutional neural network), Fast R-CNN (Fast Region-Convolutional Neural Networks, fast regional convolutional neural network), YOLOv4 (You Only Look Once 4th), YOLOv3 (You Only Look Once 3rd) and YOLOv5 (You Only Look Once 5th). It can be selected and set according to the actual situation, and the target detection neural network model is not limited in this embodiment.

[0069] In one embodiment, the target detection neural network model uses the content security policy dark network CSPdarknet53 as the backbone network, and the convolution layer of at least one residual block in the fourth residual block and the fifth residual block in the CSPdarknet53 adopts depth-separable convolution.

[0070] For example, Figure 3 , which is a schematic diagram of the structure of CSPdarknet53, which includes a convolutional layer and five residual blocks (i.e., a first residual block, a second residual block, a third residual block, a fourth residual block, and a fifth residual block). In this embodiment, the convolutional layer in the fourth residual block and / or the fifth residual block in CSPdarknet53 adopts depthwise separable convolution. Figure 3 The convolution layers in the fourth and fifth residual blocks in the figure both use depth-wise separable convolution.

[0071] Next, we introduce the parameter calculation amount of depth-separable convolution and standard convolution:

[0072] Among them, the parameter calculation amount C1 of the standard convolution can be determined by formula (1):

[0073] C1=(D K1 ) 2 M1N1(D F1 ) 2 ..........(1);

[0074] Among them, D K1 is the size of the convolution kernel of the standard convolution, M1 is the number of channels of the standard convolution, N1 is the number of convolution kernels of the standard convolution, and D F1 is the width of the input feature of the standard convolution.

[0075] The parameter calculation amount C2 of the depthwise separable convolution can be determined by formula (2):

[0076] C2=[(D K2 ) 2 +N2]M2(D F2 ) 2 ..........(2);

[0077] Among them, D K2 is the size of the convolution kernel of the depthwise separable convolution, N2 is the number of convolution kernels of the depthwise separable convolution, M2 is the number of channels of the depthwise separable convolution, and D F2 is the width of the input feature of the depthwise separable convolution.

[0078] Comparing the above two formulas, we can see that when the kernel size, number of channels, number of kernels, and width of input features of depthwise separable convolution and standard convolution are the same, the parameter calculation amount of depthwise separable convolution is greatly reduced compared with that of standard convolution, which is only 1 / 2 of the parameter calculation amount of standard convolution.

[0079] In order to improve the accuracy of detecting electric police equipment, in one embodiment, the activation function in the depthwise separable convolution is a mish activation function.

[0080] For example, Figure 4 As shown, it is a schematic diagram of the structure of the depthwise separable convolution in this embodiment, which mainly includes 3×3 depthwise convolution and 1×1 pointwise convolution. The channel is first traversed using the depthwise convolution with a convolution kernel size of 3×3, and then the channel is scaled using the pointwise convolution with a convolution kernel size of 1×1. Among them, the 3×3 depthwise convolution and the 1×1 pointwise convolution are two independent modules, and the output parts of these two modules include BN (Batch Normalization, normalization unit) and mish activation function.

[0081] Below, taking the target detection neural network model as YOLOv4 as an example, the process of using the target detection neural network model to detect vehicles in the image in this application is described in detail. Figure 5 The figure shows the structure diagram of YOLOv4, including: backbone network, SPP module (SPATIAL Pyramid Pooling), PANet network structure layer and detection layer.

[0082] Among them, the backbone network is a CSPdarknet53 network structure in this embodiment, in which the convolution layer of at least one residual block in the fourth residual block and the fifth residual block is a depth-separable convolution. The backbone network is mainly used to extract features from the video image to be detected in the road section video in this embodiment. The output of the backbone network is divided into three branches, and the size of the feature map output by each branch is different.

[0083] The SPP module uses three pooling layers of different sizes to perform pooling operations on the feature maps extracted by the backbone network. That is, 5*5, 9*9 and 13*13 pooling layers are used to perform pooling operations on the feature maps extracted by the backbone network, and then the fully connected layer is used to splice the feature maps output by the three pooling layers. After splicing, the convolution layer is used to reduce the dimension. In this way, the features obtained are richer.

[0084] The PANet network structure layer mainly performs upsampling or downsampling on the feature maps obtained from each branch, then uses the fully connected layer to concatenate the corresponding feature maps, and then uses the convolutional layer to perform dimensionality reduction. Three feature maps are obtained.

[0085] The detection layer performs vehicle detection on the three feature maps obtained in the PANet network structure layer to obtain the final recognition result.

[0086] Step 203: deleting a target vehicle from the image to obtain a target image, wherein the target vehicle is determined by the position of each vehicle in the image and the position of each vehicle in the next frame of the image, and there are duplicate vehicles with each vehicle in the next frame of the image;

[0087] In one embodiment, the target vehicle in the image is determined by:

[0088] For any vehicle in the image, perform the following steps:

[0089] According to the position of any one of the vehicles and the position of each vehicle in the next frame of the image, the distances between any one of the vehicles and each vehicle in the next frame of the image are obtained; if there is a distance among the distances that satisfies a preset distance range, it is determined that any one of the vehicles is the target vehicle in the image.

[0090] For example, Figure 6 As shown, the video images in this cycle are 3 images as an example for explanation. Among them, if the distance between the two vehicles, vehicle A in the n-1th frame image and vehicle C in the nth frame image, is within the preset distance range, then the vehicle A in the n-1th frame image is determined as the target vehicle in the n-1th frame image, and if the distance between the vehicle B in the nth frame image and the vehicle M in the n+1th frame image is within the preset range, then the vehicle B in the nth frame image is determined as the target vehicle in the nth frame image.

[0091] Step 204: Obtain the actual number of each vehicle in the current cycle according to the total number of each vehicle in each target image;

[0092] The number of each vehicle in each target image is added together to obtain the actual number of each vehicle in this cycle.

[0093] Step 205: Determine whether the recognition efficiency of the electronic police device is reduced based on the actual number and the detection number, wherein the detection number is the number of vehicles recognized by the electronic police device in this cycle.

[0094] In one embodiment, whether the recognition efficiency of the electronic police device is reduced is determined by:

[0095] According to the actual number and the detected number, the recognition efficiency of the electric police device is determined; the recognition efficiency is compared with the preset standard recognition efficiency; if the recognition efficiency is less than the standard recognition efficiency, it is determined that the recognition efficiency of the electric police device has decreased; if the recognition efficiency is not less than the standard recognition efficiency, it is determined that the recognition efficiency of the electric police device has not decreased. The recognition efficiency of the electric police device can be determined by formula (3):

[0096]

[0097] in, is the recognition efficiency of the electronic police equipment, m is the detection quantity, and N is the actual quantity.

[0098] In order to further understand the technical solution of the present disclosure, Figure 7 A detailed description may include the following steps:

[0099] Step 701: for any electric police device, periodically obtain the target video shot by the electric police device in this period;

[0100] Step 702: for any frame image in the target video, use the target detection neural network model to perform vehicle detection on the target detection area in the image to obtain the position of each vehicle;

[0101] Step 703: according to the position of the arbitrary vehicle and the positions of the vehicles in the next frame of the image, obtaining the distances between the arbitrary vehicle and the vehicles in the next frame of the image;

[0102] Step 704: If there is a distance in the distances that satisfies the preset distance range, determining that any one of the vehicles is a target vehicle in the image;

[0103] Step 705: deleting a target vehicle from the image to obtain a target image, wherein the target vehicle is determined by the position of each vehicle in the image and the position of each vehicle in the next frame of the image, and there are duplicate vehicles with each vehicle in the next frame of the image;

[0104] Step 706: Obtain the actual number of each vehicle in the current cycle according to the total number of each vehicle in each target image;

[0105] Step 707: determining the recognition efficiency of the electronic police device according to the actual number and the detected number, wherein the detected number is the number of vehicles recognized by the electronic police device in this cycle;

[0106] Step 708: Compare the recognition efficiency with a preset standard recognition efficiency to determine whether the recognition efficiency is less than the preset standard recognition efficiency. If so, execute step 709; if not, execute step 710;

[0107] Step 709: Determine that the recognition efficiency of the electronic police device is reduced;

[0108] Step 710: Determine that the recognition efficiency of the electronic police equipment has not decreased.

[0109] Based on the same disclosure concept, the method for detecting the recognition efficiency of electric police equipment as described above can also be implemented by a device for detecting the recognition efficiency of electric police equipment. The effect of the device for detecting the recognition efficiency of electric police equipment is similar to that of the above method, and will not be repeated here.

[0110] Figure 8 The present invention is a schematic diagram of the structure of an apparatus for detecting the recognition efficiency of electric police equipment according to an embodiment of the present disclosure.

[0111] like Figure 8 As shown, the apparatus 800 for detecting the recognition efficiency of electric alarm equipment disclosed herein may include an acquisition module 810 , a vehicle detection module 820 , a deletion module 830 , an actual vehicle quantity determination module 840 and a recognition efficiency determination module 850 .

[0112] An acquisition module 810 is used to periodically acquire a target video shot by any electric police device in the current period;

[0113] The vehicle detection module 820 is used to detect vehicles in the target detection area of ​​any frame image in the target video using the target detection neural network model to obtain the position of each vehicle;

[0114] A deleting module 830 is used to delete a target vehicle from the image to obtain a target image, wherein the target vehicle is determined by the position of each vehicle in the image and the position of each vehicle in the next frame of the image, and there is a duplicate vehicle with each vehicle in the next frame of the image;

[0115] The actual vehicle quantity determination module 840 is used to obtain the actual quantity of each vehicle in the current cycle according to the total quantity of each vehicle in each target image;

[0116] The recognition efficiency determination module 850 is used to determine whether the recognition efficiency of the electronic police equipment is reduced through the actual number and the detection number, wherein the detection number is the number of vehicles recognized by the electronic police equipment in this cycle.

[0117] In one embodiment, the target detection neural network model uses the content security policy dark network CSPdarknet53 as the backbone network, and the convolution layer of at least one residual block in the fourth residual block and the fifth residual block in the CSPdarknet53 adopts depth-separable convolution.

[0118] In one embodiment, the activation function in the depthwise separable convolution is a mish activation function.

[0119] In one embodiment, the apparatus further comprises:

[0120] The target vehicle determination module 860 is used to determine the target vehicle in the image by:

[0121] For any vehicle in the image, perform the following steps:

[0122] According to the position of the arbitrary vehicle and the position of each vehicle in the next frame of the image, obtaining each distance between the arbitrary vehicle and each vehicle in the next frame of the image;

[0123] If there is a distance among the distances that satisfies a preset distance range, it is determined that any one of the vehicles is a target vehicle in the image.

[0124] In one embodiment, the recognition efficiency determination module 850 is specifically configured to:

[0125] Determining the recognition efficiency of the electric police device according to the actual number and the detected number;

[0126] Comparing the recognition efficiency with a preset standard recognition efficiency;

[0127] If the recognition efficiency is less than the standard recognition efficiency, it is determined that the recognition efficiency of the electric police device is reduced;

[0128] If the recognition efficiency is not less than the standard recognition efficiency, it is determined that the recognition efficiency of the electronic police equipment has not decreased.

[0129] After introducing a method and apparatus for detecting recognition efficiency of an electric alarm device according to an exemplary embodiment of the present disclosure, next, an electronic device according to another exemplary embodiment of the present disclosure is introduced.

[0130] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented as systems, methods or program products. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to herein as "circuits", "modules" or "systems".

[0131] In some possible implementations, the electronic device according to the present disclosure may include at least one processor and at least one computer storage medium. The computer storage medium stores program code, and when the program code is executed by the processor, the processor executes the steps in the method for detecting the recognition efficiency of the electric alarm device according to various exemplary embodiments of the present disclosure described above in this specification. For example, the processor may execute the following steps: Figure 2Steps 201-205 shown in .

[0132] Refer to the following Fig. 9 The electronic device 900 according to this embodiment of the present disclosure is described. Fig. 9 The electronic device 900 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0133] like Fig. 9 As shown, the electronic device 900 is in the form of a general electronic device. The components of the electronic device 900 may include but are not limited to: at least one processor 901, at least one computer storage medium 902, and a bus 903 connecting different system components (including the computer storage medium 902 and the processor 901).

[0134] Bus 903 represents one or more of several types of bus structures, including a computer storage media bus or computer storage media controller, a peripheral bus, a processor, or a local bus using any of a variety of bus architectures.

[0135] The computer storage media 902 may include readable media in the form of volatile computer storage media, such as random access computer storage media (RAM) 921 and / or cache storage media 922 , and may further include read-only computer storage media (ROM) 923 .

[0136] The computer storage medium 902 may also include a program / utility 925 having a set (at least one) of program modules 924, such program modules 924 including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include an implementation of a network environment.

[0137] The electronic device 900 may also communicate with one or more external devices 904 (e.g., keyboards, pointing devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 900, and / or communicate with any device that enables the electronic device 900 to communicate with one or more other electronic devices (e.g., routers, modems, etc.). This communication may be performed through an input / output (I / O) interface 905. In addition, the electronic device 900 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter 906. As shown, the network adapter 906 communicates with other modules for the electronic device 900 through a bus 903. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 900, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0138] In some possible implementations, various aspects of a method for detecting the recognition efficiency of electronic police equipment provided by the present disclosure may also be implemented in the form of a program product, which includes program code. When the program product is run on a computer device, the program code is used to enable the computer device to execute the steps of the method for detecting the recognition efficiency of electronic police equipment according to various exemplary embodiments of the present disclosure described above in this specification.

[0139] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access computer storage medium (RAM), a read-only computer storage medium (ROM), an erasable programmable read-only computer storage medium (EPROM or flash memory), an optical fiber, a portable compact disk read-only computer storage medium (CD-ROM), an optical computer storage medium, a magnetic computer storage medium, or any suitable combination of the above.

[0140] The program product for detecting the recognition efficiency of electric alarm equipment of the embodiment of the present disclosure can adopt a portable compact disk read-only computer storage medium (CD-ROM) and include program code, and can be run on an electronic device. However, the program product of the present disclosure is not limited to this. In this document, the readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, device or device.

[0141] The readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, wherein the readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable signal medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0142] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0143] Program code for performing the disclosed operations may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user electronic device, partially on the user device, as a separate software package, partially on the user electronic device and partially on a remote electronic device, or entirely on a remote electronic device or server. In the case of a remote electronic device, the remote electronic device may be connected to the user electronic device via any type of network including a local area network (LAN) or a wide area network (WAN), or may be connected to an external electronic device (e.g., via the Internet using an Internet service provider).

[0144] It should be noted that although several modules of the device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules described above can be embodied in one module. Conversely, the features and functions of one module described above can be further divided into multiple modules to be embodied.

[0145] In addition, although the operations of the disclosed method are described in a specific order in the drawings, this does not require or imply that the operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

[0146] Those skilled in the art will appreciate that the embodiments of the present disclosure may be provided as methods, systems, or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present disclosure may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk computer storage media, CD-ROM, optical computer storage media, etc.) containing computer-usable program codes.

[0147] The present disclosure is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the present disclosure. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0148] These computer program instructions may also be stored in a computer-readable computer storage medium that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable computer storage medium produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0149] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0150] Obviously, those skilled in the art can make various changes and modifications to the present disclosure without departing from the spirit and scope of the present disclosure. Thus, if these modifications and variations of the present disclosure fall within the scope of the claims of the present disclosure and their equivalents, the present disclosure is also intended to include these modifications and variations.

Claims

1. An electronic device, characterized in that: The device comprises a processor and a storage unit, wherein: The storage unit is configured to store the video captured by the electronic police equipment; The processor is configured to: For any electric police device, periodically obtain the target video shot by the electric police device in this period; For any frame of the target video, use the target detection neural network model to perform vehicle detection on the target detection area in the image to obtain the position of each vehicle; The target vehicle is deleted from the image to obtain a target image, wherein the target vehicle in the image is determined in the following manner: for any vehicle in the image, according to the position of the any vehicle and the position of each vehicle in the next frame of the image, the distances between the any vehicle and each vehicle in the next frame of the image are obtained; if there is a distance in the distances that satisfies a preset distance range, the any vehicle is determined to be the target vehicle in the image; According to the total number of each vehicle in each target image, the actual number of each vehicle in this cycle is obtained; Whether the recognition efficiency of the electric police equipment is reduced is determined by the actual number and the detection number, wherein the detection number is the number of vehicles recognized by the electric police equipment in this cycle.

2. The electronic device according to claim 1, characterized in that: The target detection neural network model uses the content security policy dark network CSPdarknet53 as the backbone network, and the convolution layer of at least one residual block in the fourth residual block and the fifth residual block in the CSPdarknet53 adopts depth-separable convolution.

3. The electronic device according to claim 2, characterized in that: The activation function in the depthwise separable convolution is the mish activation function.

4. The electronic device according to any one of claims 1 to 3, characterized in that: The processor performs the step of determining whether the recognition efficiency of the electric police device is reduced by using the actual number and the detected number, and is specifically configured as follows: Determining the recognition efficiency of the electric police device according to the actual number and the detected number; Comparing the recognition efficiency with a preset standard recognition efficiency; If the recognition efficiency is less than the standard recognition efficiency, it is determined that the recognition efficiency of the electric police device is reduced; If the recognition efficiency is not less than the standard recognition efficiency, it is determined that the recognition efficiency of the electronic police equipment has not decreased.

5. A method for detecting the recognition efficiency of electric police equipment, characterized in that: The method comprises: For any electric police device, periodically obtain the target video shot by the electric police device in this period; For any frame of the target video, use the target detection neural network model to perform vehicle detection on the target detection area in the image to obtain the position of each vehicle; The target vehicle is deleted from the image to obtain a target image, wherein the target vehicle in the image is determined in the following manner: for any vehicle in the image, according to the position of the any vehicle and the position of each vehicle in the next frame of the image, the distances between the any vehicle and each vehicle in the next frame of the image are obtained; if there is a distance in the distances that satisfies a preset distance range, the any vehicle is determined to be the target vehicle in the image; According to the total number of each vehicle in each target image, the actual number of each vehicle in this cycle is obtained; Whether the recognition efficiency of the electric police equipment is reduced is determined by the actual number and the detection number, wherein the detection number is the number of vehicles recognized by the electric police equipment in this cycle.

6. The method according to claim 5, characterized in that The target detection neural network model uses the content security policy dark network CSPdarknet53 as the backbone network, and the convolution layer of at least one residual block in the fourth residual block and the fifth residual block in the CSPdarknet53 adopts depth-separable convolution.

7. The method according to claim 6, characterized in that The activation function in the depthwise separable convolution is the mish activation function.

8. The method according to any one of claims 5 to 7, characterized in that: The determining whether the recognition efficiency of the electric police device is reduced by using the actual number and the detected number includes: Determining the recognition efficiency of the electric police device according to the actual number and the detected number; Comparing the recognition efficiency with a preset standard recognition efficiency; If the recognition efficiency is less than the standard recognition efficiency, it is determined that the recognition efficiency of the electric police device is reduced; If the recognition efficiency is not less than the standard recognition efficiency, it is determined that the recognition efficiency of the electronic police equipment has not decreased.

Images