Box door identification method and device, electronic equipment, storage medium and program product

Abstract

The embodiment of the invention provides a box door recognition method and device, electronic equipment, a storage medium and a program product, and relates to the technical field of image recognition. The method comprises the following steps: acquiring an identification request; collecting a to-be-recognized image according to the recognition request; the to-be-recognized image comprises the box door. According to a preset target recognition model, performing state detection on the box door in the to-be-recognized image to obtain state information; the state information represents that the box door is in an open state or a closed state. Sending the state information to a preset control system; wherein the state information is used for generating prompt information. The method is used for achieving the effect of improving the accuracy of opening and closing detection of the container door.

Description

Technical Field

[0001] The present application relates to the field of image recognition technology, and in particular to a box door recognition method, device, electronic device, storage medium and program product. Background Art

[0002] At present, for the containers in the yard, during the handover, storage and custody, it is necessary to ensure that the container doors are closed. The door opening and closing detection technology is needed to detect the opening and closing status of the container doors, but the door opening and closing detection technology is relatively difficult.

[0003] In the prior art, the main existing solution is based on radar, which scans both sides of the container and determines the open or closed state of the container door based on the prior inspection box ROI area.

[0004] However, in the prior art, since the door opening and closing status is determined by radar, the radar-based solution has many disadvantages. For example, there is a lot of interference. For example, if there is a bulge on one side of the container, it is very easy to cause a false alarm; or the radar scanning angle is limited, and it is impossible to effectively monitor the door with an angle less than 10°; which leads to the low accuracy of the existing door opening and closing detection. Summary of the invention

[0005] The embodiments of the present application provide a container door recognition method, device, electronic device, storage medium and program product, which are used to improve the accuracy of opening and closing detection of container doors.

[0006] In a first aspect, an embodiment of the present application provides a method for identifying a door, comprising:

[0007] Obtaining an identification request; wherein the identification request is used to identify the open or closed state of the door;

[0008] According to the recognition request, an image to be recognized is collected; wherein the image to be recognized includes a door;

[0009] According to a preset target recognition model, the state of the box door in the image to be recognized is detected to obtain state information; wherein the preset target recognition model is obtained by training based on a labeled box door data set; the state information indicates that the box door is in an open state or in a closed state;

[0010] The status information is sent to a preset control system; wherein the status information is used to generate prompt information.

[0011] In a possible implementation, the state detection of the door in the image to be identified is performed according to a preset target recognition model to obtain the state information, including:

[0012] According to a preset target recognition model, an image of a preset range area where the door of the box is located in the image to be recognized is identified;

[0013] Extracting door features in the image of the preset range area;

[0014] The state of the door features in the preset range area image is detected to obtain state information.

[0015] In a possible implementation, performing state detection on the door features in the preset range area image to obtain state information includes:

[0016] If the preset target recognition model is a target detection model, then according to the target detection model, the state detection of the door features in the preset range area image is performed to obtain the opening and closing angle of the door;

[0017] If the opening and closing angle is greater than a preset angle threshold, state information indicating that the door is in an open state is obtained;

[0018] If the opening and closing angle is less than or equal to the angle threshold, state information indicating that the door is in a closed state is obtained.

[0019] In a possible implementation, performing state detection on the door features in the preset range area image to obtain state information includes:

[0020] If the preset target recognition model is an image classification model, the preset range area image is input into the image classification model, and the state of the door features in the preset range area image is detected to obtain classification result information; wherein the classification result information indicates that the category of the door is an open type or a closed type;

[0021] According to the classification result information, state information is determined.

[0022] In a possible implementation manner, collecting the image to be recognized according to the recognition request includes:

[0023] Based on the identification request, determining a pre-connected camera network, a preset start trajectory step, and a preset end trajectory step;

[0024] According to the camera network, the preset start trajectory step and the preset end trajectory step, the image video is collected; wherein the camera is located at the midpoint between the two side spreaders;

[0025] The image video is parsed to obtain multiple frames of images to be recognized.

[0026] In a possible implementation, the method further includes:

[0027] Acquire multiple historical images and videos; and pre-process and analyze the historical images and videos to obtain multiple frames of historical images;

[0028] In response to the labeling operation on the multiple frames of historical images, obtaining a training data set;

[0029] Based on the training data set, the target recognition model is trained until the target recognition model converges; wherein the target recognition model is a target detection model, or an image classification model.

[0030] In one possible implementation,

[0031] In a second aspect, an embodiment of the present application provides a box door recognition device, comprising:

[0032] An acquisition module, used to acquire an identification request; wherein the identification request is used to identify the open or closed state of the door;

[0033] A collection module, used for collecting the image to be recognized according to the recognition request; wherein the image to be recognized includes a door;

[0034] A detection module, used to perform state detection on the door in the image to be identified according to a preset target recognition model to obtain state information; wherein the preset target recognition model is obtained by training based on a labeled door data set; and the state information indicates that the door is in an open state or a closed state;

[0035] A sending module is used to send the status information to a preset control system; wherein the status information is used to generate prompt information.

[0036] In a possible implementation, the detection module includes:

[0037] A recognition unit, used to recognize a preset range area image where the door of the box is located in the image to be recognized according to a preset target recognition model;

[0038] An extraction unit, used for extracting door features in the image of the preset range area;

[0039] The detection unit is used to perform status detection on the door features in the preset range area image to obtain status information.

[0040] In a possible implementation manner, the detection unit is specifically used to:

[0041] If the preset target recognition model is a target detection model, then according to the target detection model, the state detection of the door features in the preset range area image is performed to obtain the opening and closing angle of the door;

[0042] If the opening and closing angle is greater than a preset angle threshold, state information indicating that the door is in an open state is obtained;

[0043] If the opening and closing angle is less than or equal to the angle threshold, state information indicating that the door is in a closed state is obtained.

[0044] In a possible implementation manner, the detection unit is specifically used to:

[0045] If the preset target recognition model is an image classification model, the preset range area image is input into the image classification model, and the state of the door features in the preset range area image is detected to obtain classification result information; wherein the classification result information indicates that the category of the door is an open type or a closed type;

[0046] According to the classification result information, state information is determined.

[0047] In a possible implementation manner, the acquisition module is specifically used to:

[0048] Based on the identification request, determining a pre-connected camera network, a preset start trajectory step, and a preset end trajectory step;

[0049] According to the camera network, the preset start trajectory step and the preset end trajectory step, the image video is collected; wherein the camera is located at the midpoint between the two side spreaders;

[0050] The image video is parsed to obtain multiple frames of images to be recognized.

[0051] In a possible implementation manner, the device is further specifically used for:

[0052] Acquire multiple historical images and videos; and pre-process and analyze the historical images and videos to obtain multiple frames of historical images;

[0053] In response to the labeling operation on the multiple frames of historical images, obtaining a training data set;

[0054] Based on the training data set, the target recognition model is trained until the target recognition model converges; wherein the target recognition model is a target detection model, or an image classification model.

[0055] In a third aspect, an embodiment of the present application provides an electronic device, including: a memory, a processor;

[0056] The memory stores computer-executable instructions;

[0057] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementations of the first aspect.

[0058] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the first aspect above and / or various possible implementations of the first aspect.

[0059] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above first aspect and / or various possible implementation methods of the first aspect.

[0060] The container door recognition method, device, electronic device, storage medium and program product provided in the embodiments of the present application obtain a recognition request; wherein the recognition request is used to identify the open and closed state of the container door. According to the recognition request, an image to be recognized is collected; wherein the image to be recognized includes the container door. According to a preset target recognition model, the state of the container door in the image to be recognized is detected to obtain state information; wherein the preset target recognition model is obtained by training based on a labeled container door data set; the state information represents that the container door is in an open state, or in a closed state. The state information is sent to a preset control system; wherein the state information is used to generate prompt information. In this solution, the closed state of the container door is detected by a preset target recognition model, which can improve stability and robustness, does not require a cumbersome calibration process, and can ensure a high accuracy rate, so as to achieve the effect of improving the accuracy of the opening and closing detection of the container door. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0062] Figure 1 A schematic diagram of a method for identifying a door provided in an embodiment of the present application Figure 1 ;

[0063] Figure 2 Schematic diagram of another method for identifying a door provided in an embodiment of the present application Figure 2 ;

[0064] Figure 3 Schematic diagram of another method for identifying a door provided in an embodiment of the present application Figure 3 ;

[0065] Figure 4 A schematic diagram of the structure of a door identification device provided in an embodiment of the present application;

[0066] Figure 5 A schematic diagram of the structure of another cabinet door identification device provided in an embodiment of the present application;

[0067] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0068] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0069] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0070] At present, for the containers in the yard, during the handover, storage and custody, it is necessary to ensure that the container doors are closed. The door opening and closing detection technology is needed to detect the opening and closing status of the container doors, but the door opening and closing detection technology is relatively difficult.

[0071] In one example, the existing main solution is based on radar, scanning both sides of the container, and determining the door opening and closing status based on the prior inspection box ROI area. However, in the existing technology, since the door opening and closing status is determined based on radar, the radar-based solution has many disadvantages, such as a lot of interference. For example, if there is a bulge on one side of the container, it is very easy to cause a false alarm; or the radar scanning angle is limited, and it cannot effectively monitor the door with an angle less than 10°; which leads to the low accuracy of the existing door opening and closing detection.

[0072] In one example, based on radar, the two sides of the container are scanned, and the opening and closing status of the container door is determined according to the prior inspection box ROI area. The ROI area needs to be calibrated in advance. The calibration work is relatively cumbersome, resulting in low calibration efficiency.

[0073] In combination with the above scenario, it can be seen that in the prior art, the door opening and closing status is determined by radar, but there is a technical problem that the accuracy of the existing door opening and closing detection is low.

[0074] The door recognition method provided in the present application detects the closed state of the door through a preset target recognition model, which can improve stability and robustness, does not require a cumbersome calibration process, and can ensure a high accuracy rate, thereby solving the technical problem of low accuracy in opening and closing detection of container doors.

[0075] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0076] Figure 1 A schematic diagram of a method for identifying a door provided in this application Figure 1 ,like Figure 1 As shown, the method includes:

[0077] S101, obtaining an identification request; wherein the identification request is used to identify the open or closed state of a door.

[0078] For example, the execution subject of this embodiment may be an electronic device, or a terminal device, or a sensitive word data processing device or device, or other devices or equipment that can execute this embodiment, and there is no limitation on this. In this embodiment, the execution subject is introduced as an electronic device.

[0079] First, for the door of the container in the yard, an identification request is obtained, and the identification request is used to identify the open or closed state of the door.

[0080] S102. According to the recognition request, collect the image to be recognized; wherein the image to be recognized includes the door of the box.

[0081] Exemplarily, at least one camera is installed at the midpoint between the two side spreaders so that the camera is aimed at the area where the box door is located. Based on the recognition request, the camera collects the image to be recognized, and the image to be recognized includes the box door. For example, when it is determined that the current state is in the box picking state and / or the box placing state, the camera collects the image to be recognized. Therefore, the opening and closing of the box door can be clearly observed through the camera.

[0082] S103. According to a preset target recognition model, a state detection is performed on the door in the image to be recognized to obtain state information; wherein the preset target recognition model is trained based on a labeled door data set; and the state information indicates whether the door is in an open state or a closed state.

[0083] Exemplarily, the preset target recognition model is obtained based on the labeled door data set and deep learning network training. The target recognition model is a model for identifying the opening and closing of the door. For example, the target recognition model can be a target detection model, or a classification recognition model, etc., which is not limited. In this step, according to the preset target recognition model, the state of the door in the image to be recognized is detected to obtain state information, and the state information indicates that the door is in an open state or in a closed state.

[0084] S104, sending the status information to a preset control system; wherein the status information is used to generate prompt information.

[0085] Exemplarily, the state information is sent to a preset control system. In the control system, if the state information indicates that the door is in an open state, a prompt message can be generated, thereby promptly reminding the user.

[0086] The container door recognition method provided in the embodiment of the present application obtains a recognition request; wherein the recognition request is used to identify the open and closed state of the container door. According to the recognition request, an image to be recognized is collected; wherein the image to be recognized includes the container door. According to a preset target recognition model, the state of the container door in the image to be recognized is detected to obtain state information; wherein the preset target recognition model is obtained by training based on a labeled container door data set; the state information represents that the container door is in an open state, or in a closed state. The state information is sent to a preset control system; wherein the state information is used to generate prompt information. In this solution, the closed state of the container door is detected by the target recognition model obtained by deep learning network training, which can improve stability and robustness, does not require a cumbersome calibration process, and can ensure a high accuracy rate, so as to achieve the effect of improving the accuracy of the opening and closing detection of the container door.

[0087] Figure 2 A schematic diagram of a method for identifying a door provided in this application Figure 2 ,like Figure 2 As shown, in this embodiment Figure 1 Based on the embodiment, a box door identification method is described in detail, and the method includes:

[0088] S201, obtaining a plurality of historical images and videos; and preprocessing and analyzing the historical images and videos to obtain a plurality of frames of historical images.

[0089] Exemplarily, a camera is installed at the midpoint between the two side spreaders, and the electronic device can be connected to the preset camera network, and multiple historical image videos are collected through the camera, and the historical image videos are preprocessed and analyzed to obtain multiple frames of historical images. Specifically, the data is collected using a preset image acquisition software, and the start track step and the end track step of the door opening and closing detection are determined by programming a preset programmable logic controller (PLC), and the starting node of the video recording is determined by the starting track step, and the end node of the video recording is determined by the end track step. According to the starting node and the end node, the image video of the yard door is recorded. Then the historical image video is analyzed frame by frame to obtain the door orientation video, and the original data set is obtained according to the door orientation video. The original data set is the multiple frames of historical images in the door orientation video. Further, by performing preprocessing such as data cleaning on the historical image video, the pictures that do not include the door opening and closing working conditions in the data set are eliminated, and then the data set to be labeled is obtained.

[0090] S202: In response to the labeling operation on multiple frames of historical images, a training data set is obtained.

[0091] Exemplarily, in response to a labeling operation on multiple frames of historical images, a training data set is obtained, wherein the labeling operation is a labeling performed on a door. For example, a preset labeling application is called, and in response to the labeling operation in the labeling application, the training data set is obtained.

[0092] S203. Based on the training data set, train the target recognition model until the target recognition model converges; wherein the target recognition model is a target detection model, or an image classification model.

[0093] Exemplarily, the electronic device can set the model parameters of the target recognition model, and the model parameters can be preset parameters or custom parameters, which are not limited. Based on the acquired training data set setting, the target recognition model is trained until the target recognition model converges and the training is stopped. The target recognition model obtained at this time is the final target recognition model, and the target recognition model can be a target detection model or an image classification model. Finally, the trained target detection model or image classification model is used to build an inference framework to achieve the purpose of inputting an image to obtain a detection result.

[0094] S204, obtaining an identification request; wherein the identification request is used to identify the open or closed state of the door.

[0095] For example, this step can be referred to Figure 1 Step 101 in the above is not described in detail.

[0096] S205. According to the recognition request, collect the image to be recognized; wherein the image to be recognized includes the door of the box.

[0097] In one example, step S205 includes: based on the recognition request, determining a pre-connected camera network, a preset start trajectory step, and a preset end trajectory step; collecting image videos according to the camera network, the preset start trajectory step, and the preset end trajectory step; wherein the camera is located at the midpoint between the two side hoists; parsing the image video to obtain multiple frames of images to be recognized.

[0098] Exemplarily, the electronic device can install at least one camera at the midpoint between the two side slings so that the camera is aimed at the area where the box door is located. Based on the recognition request, the pre-connected camera network, the preset start trajectory step and the preset end trajectory step, the camera collects the image video, and the image video is parsed to obtain multiple frames of images to be recognized, and the images to be recognized include the box door. For example, when it is determined that the current state is in the box picking state and / or the box placing state, the camera collects the image video, and the image video is parsed to obtain multiple frames of images to be recognized.

[0099] S206: Identify the preset range area image where the door of the box is located in the image to be identified according to the preset target recognition model.

[0100] Exemplarily, the electronic device can identify the preset range area image where the door of the box is located in the image to be identified according to the preset target recognition model, wherein the area occupied by the preset range area image is the preset range. Specifically, the door of the box can be taken as the center position, and the preset range can be determined with the door as the center, and then the preset range area image corresponding to the preset range can be determined.

[0101] S207: extracting door features in the preset range area image.

[0102] Exemplarily, the electronic device may extract door features in an image of a preset range area according to a preset target recognition model.

[0103] S208: Perform status detection on the door features in the preset range area image to obtain status information.

[0104] Step 208 includes two implementations:

[0105] The first implementation method of step 208: If the preset target recognition model is a target detection model, then according to the target detection model, the state detection of the door features in the preset range area image is performed to obtain the opening and closing angle of the door; if the opening and closing angle is greater than the preset angle threshold, the state information representing that the door is in the open state is obtained; if the opening and closing angle is less than or equal to the angle threshold, the state information representing that the door is in the closed state is obtained.

[0106] A second implementation method of step 208: If the preset target recognition model is an image classification model, the preset range area image is input into the image classification model, and the state detection is performed on the box door features in the preset range area image to obtain classification result information; wherein the classification result information characterizes the category of the box door as an open type, or a closed type; and the state information is determined based on the classification result information.

[0107] Exemplarily, the preset target recognition model may be a target detection model, or a classification recognition model, etc., which is not limited. In this step, according to the target detection model, the state of the box door in the image to be recognized is detected to obtain state information; or, according to the classification recognition model, the state of the box door in the image to be recognized is detected to obtain state information; wherein the state information indicates that the box door is in an open state, or in a closed state.

[0108] In the first implementation, if the preset target recognition model is a target detection model, the state detection is performed on the door features in the preset range area image according to the target detection model to obtain the opening and closing angle of the door. The opening and closing angle is compared with the preset angle threshold. If it is determined that the opening and closing angle is greater than the preset angle threshold, the state information obtained at this time indicates that the door is in an open state; if it is determined that the opening and closing angle is less than or equal to the angle threshold, the state information obtained at this time indicates that the door is in a closed state.

[0109] In the second implementation, if the preset target recognition model is an image classification model, the preset range area image is input into the image classification model, the state of the door features in the preset range area image is detected and classified, and classification result information is obtained, and the classification result information indicates that the category of the door is an open type or a closed type. If the classification result information indicates that the category of the door is an open type, it is determined that the state information indicates that the door is in an open state, and if the classification result information indicates that the category of the door is a closed type, it is determined that the state information indicates that the door is in a closed state.

[0110] S209, sending the status information to a preset control system; wherein the status information is used to generate prompt information.

[0111] For example, the electronic device can transmit the detected state information to the control system through a preset interface, and the state information is used to generate prompt information to further implement the application of the state information. Among them, the preset interface can be a POST interface, etc., which is not limited to this.

[0112] The door recognition method provided in the embodiment of the present application obtains multiple historical image videos; and pre-processes and analyzes the historical image videos to obtain multiple frames of historical images. In response to the annotation operation on the multiple frames of historical images, a training data set is obtained. Based on the training data set, the target recognition model is trained until the target recognition model converges; wherein the target recognition model is a target detection model, or an image classification model. Obtain a recognition request; wherein the recognition request is used to identify the opening and closing state of the door. According to the recognition request, the image to be recognized is collected; wherein the image to be recognized includes the door. According to the preset target recognition model, the preset range area image where the door in the image to be recognized is located is recognized. The door features in the preset range area image are extracted. The state of the door features in the preset range area image is detected to obtain state information. The state information is sent to the preset control system; wherein the state information is used to generate prompt information. In this solution, the target recognition model obtained by deep learning network training detects the closed state of the door, which can improve stability and robustness, does not require a cumbersome calibration process, and can ensure a high accuracy rate, so as to achieve the effect of improving the accuracy of the opening and closing detection of the container door.

[0113] Figure 3 A schematic diagram of a method for identifying a door provided in this application Figure 3 ,like Figure 3 As shown, the target recognition model takes the target detection model as an example, including: start; the system collects pictures, the system refers to the system of the electronic device; judge whether it is a valid image, that is, judge whether the image includes the box door to be identified, if it is included, it is a valid image, if not, it is an invalid image; if it is an invalid image, continue to collect images until a valid image is collected; if it is a valid image, enter the target detection model for detection, and judge whether the box door is open; if it is in the open state, upload it to the preset control system, PLC error, manual intervention, etc.; end; if it is in the closed state, continue to collect pictures and detect. It should be noted that the detection process of the target classification model is the same as that of the target detection model. The detection process of the target classification model can refer to the above target detection model, which will not be repeated here.

[0114] Figure 4 A schematic diagram of the structure of a door identification device provided in this application, such as Figure 4 As shown, the door identification device 30 provided in this embodiment includes:

[0115] The acquisition module 31 is used to acquire an identification request, wherein the identification request is used to identify the open or closed state of the door.

[0116] The acquisition module 32 is used to acquire the image to be identified according to the identification request; wherein the image to be identified includes the door.

[0117] The detection module 33 is used to perform state detection on the door in the image to be identified according to a preset target recognition model to obtain state information; wherein the preset target recognition model is trained based on a labeled door data set; the state information represents whether the door is in an open state or a closed state.

[0118] The sending module 34 is used to send the status information to a preset control system; wherein the status information is used to generate prompt information.

[0119] Figure 5 A schematic diagram of the structure of another door identification device provided in an embodiment of the present application, Figure 4 Based on the embodiment shown, Figure 5 As shown, the detection module 33 includes:

[0120] The recognition unit 331 is used to recognize a preset range area image where the door of the box is located in the image to be recognized according to a preset target recognition model.

[0121] The extraction unit 332 is used to extract the door features in the preset range area image.

[0122] The detection unit 333 is used to perform status detection on the door features in the preset range area image to obtain status information.

[0123] In a possible implementation, the detection unit 333 is specifically configured to:

[0124] If the preset target recognition model is a target detection model, the state of the door features in the preset range area image is detected according to the target detection model to obtain the opening and closing angle of the door.

[0125] If the opening and closing angle is greater than a preset angle threshold, state information indicating that the door is in an open state is obtained.

[0126] If the opening and closing angle is less than or equal to the angle threshold, state information indicating that the door is in a closed state is obtained.

[0127] In a possible implementation, the detection unit 333 is specifically configured to:

[0128] If the preset target recognition model is an image classification model, the preset range area image is input into the image classification model, and the state of the box door features in the preset range area image is detected to obtain classification result information; wherein the classification result information characterizes the category of the box door as an open type, or a closed type.

[0129] According to the classification result information, the status information is determined.

[0130] In a possible implementation, the acquisition module 32 is specifically configured to:

[0131] Based on the recognition request, a pre-connected camera network, a preset start trajectory step, and a preset end trajectory step are determined.

[0132] Image videos are collected according to the camera network, the preset start trajectory step, and the preset end trajectory step; wherein the camera is located at the midpoint between the two side spreaders.

[0133] The image video is parsed to obtain multiple frames of images to be recognized.

[0134] In a possible implementation, the device is further specifically used for:

[0135] Acquire multiple historical image videos; and pre-process and parse the historical image videos to obtain multiple frames of historical images.

[0136] In response to a labeling operation on multiple frames of historical images, a training data set is obtained.

[0137] Based on the training data set, the target recognition model is trained until the target recognition model converges; wherein the target recognition model is a target detection model, or an image classification model.

[0138] The device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and this embodiment will not be described in detail here.

[0139] Figure 6 This is a schematic diagram of the structure of the electronic device provided in this application. Figure 6 As shown, the electronic device 40 provided in this embodiment includes: at least one processor 401 and a memory 402. Optionally, the device 40 also includes a communication component 403. The processor 401, the memory 402 and the communication component 403 are connected via a bus 404.

[0140] In a specific implementation process, at least one processor 401 executes the computer-executable instructions stored in the memory 402, so that at least one processor 401 executes the above method.

[0141] The specific implementation process of the processor 401 can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.

[0142] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the invention may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0143] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (Non-volatile Memory, NVM), such as at least one disk memory.

[0144] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of the present application is not limited to only one bus or one type of bus.

[0145] The present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0146] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.

[0147] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0148] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (Application Specific Integrated Circuits, referred to as: ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0149] The division of units is only a logical function division, and there may be other divisions in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0150] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0151] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0152] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0153] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0154] Finally, it should be noted that those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention, which follow the general principles of the present invention and include common knowledge or customary technical means in the art not disclosed by the present invention, are not limited to the precise structure described above and shown in the drawings, and may be modified and changed in various ways without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A method for identifying a door, characterized in that: include: Obtaining an identification request; wherein the identification request is used to identify the open or closed state of the door; According to the recognition request, an image to be recognized is collected; wherein the image to be recognized includes a door; According to a preset target recognition model, the state of the box door in the image to be recognized is detected to obtain state information; wherein the preset target recognition model is obtained by training based on a labeled box door data set; the state information indicates that the box door is in an open state or in a closed state; The status information is sent to a preset control system; wherein the status information is used to generate prompt information.

2. The method according to claim 1, characterized in that The method of performing state detection on the door in the image to be identified according to the preset target recognition model to obtain state information includes: According to a preset target recognition model, an image of a preset range area where the door of the box is located in the image to be recognized is identified; Extracting door features in the image of the preset range area; The state of the door features in the preset range area image is detected to obtain state information.

3. The method according to claim 2, characterized in that Performing state detection on the door features in the preset range area image to obtain state information includes: If the preset target recognition model is a target detection model, then according to the target detection model, the state detection of the door features in the preset range area image is performed to obtain the opening and closing angle of the door; If the opening and closing angle is greater than a preset angle threshold, state information indicating that the door is in an open state is obtained; If the opening and closing angle is less than or equal to the angle threshold, state information indicating that the door is in a closed state is obtained.

4. The method according to claim 2, characterized in that: Performing state detection on the door features in the preset range area image to obtain state information includes: If the preset target recognition model is an image classification model, the preset range area image is input into the image classification model, and the state of the door features in the preset range area image is detected to obtain classification result information; wherein the classification result information indicates that the category of the door is an open type or a closed type; According to the classification result information, state information is determined.

5. The method according to claim 1, characterized in that The step of collecting the image to be identified according to the identification request includes: Based on the identification request, determining a pre-connected camera network, a preset start trajectory step, and a preset end trajectory step; According to the camera network, the preset start trajectory step and the preset end trajectory step, the image video is collected; wherein the camera is located at the midpoint between the two side spreaders; The image video is parsed to obtain multiple frames of images to be recognized.

6. The method according to any one of claims 1 to 5, characterized in that: The method further comprises: Acquire multiple historical images and videos; and pre-process and analyze the historical images and videos to obtain multiple frames of historical images; In response to the labeling operation on the multiple frames of historical images, obtaining a training data set; Based on the training data set, the target recognition model is trained until the target recognition model converges; wherein the target recognition model is a target detection model, or an image classification model.

7. A door identification device, characterized in that: include: An acquisition module, used to acquire an identification request; wherein the identification request is used to identify the open or closed state of the door; A collection module, used for collecting the image to be recognized according to the recognition request; wherein the image to be recognized includes a door; A detection module, used to perform state detection on the door in the image to be identified according to a preset target recognition model to obtain state information; wherein the preset target recognition model is obtained by training based on a labeled door data set; and the state information indicates that the door is in an open state or a closed state; A sending module is used to send the status information to a preset control system; wherein the status information is used to generate prompt information.

8. An electronic device, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 6 when executed by a processor.

10. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images