A decentralized image recognition method and system for urban rail transit security inspection

Abstract

The embodiment of the present application discloses a decentralized image judgment method and system for urban rail transit security inspection, which receives in real time a scanned image of an inspected object collected by a security inspection device and an intelligent image recognition result of the scanned image by an intelligent image recognition device; packages the received scanned image and the corresponding intelligent image recognition result to form an image judgment task, which carries the identification information of the security inspection device; assigns the image judgment task to an idle image judgment station or re-inspection station, wherein each train station is provided with at least one image judgment station and at least one security inspection point, and each security inspection point is provided with at least one security inspection device and at least one re-inspection station; receives the image judgment result fed back by the image judgment station or re-inspection station, which carries the identification information of the security inspection device; and sends the image judgment result to the re-inspection station of the security inspection device corresponding to the identification information of the security inspection device; the problem of insufficient human resources for image judgment can be solved, and the utilization rate of personnel in image judgment work and the efficiency of image judgment work can be improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of urban rail transit security inspection, and in particular to a method and system for decentralized image recognition for urban rail transit security inspection. Background Art

[0002] To ensure public security along urban rail transit lines and at stations, existing urban rail transit stations typically have X-ray baggage inspection machines installed at security checkpoints at station entrances. Each station installs one or more X-ray baggage inspection machines at each security checkpoint, depending on passenger volume. The X-ray baggage inspection machine uses a conveyor belt to transport the baggage to be inspected into a crawler-type tunnel. Once the baggage enters the tunnel, an X-ray source emits an X-ray beam, using line scan imaging technology. As the conveyor belt rotates, the beam gradually penetrates the baggage and is transmitted to an X-ray detection box. This gradually forms an X-ray perspective image of the baggage on the X-ray detection box, generating an X-ray scan image. This image is displayed on a monitor, and then manually interpreted by an interpreter to determine whether any prohibited items are present.

[0003] However, since the existing X-ray baggage inspection machines are separately installed at each security checkpoint in each station, except for a few stations with large passenger flow that are basically in a state of large passenger flow all day long, most stations are in a non-peak state most of the day except for peak hours in the morning and evening, and the passenger flow entering the stations is relatively small. Even in some remote suburban stations, the passenger flow entering the stations is relatively small during peak hours in the morning and evening. This leads to an uneven number and distribution of X-ray scan images at security checkpoints in various stations in the entire urban rail transit network. Each X-ray baggage inspection machine at each security checkpoint is equipped with an interpreter around the clock, which leads to low work capacity of interpreters in different operating periods at most stations, resulting in a huge waste of manpower costs for subway operations.

[0004] To address these issues, existing technologies have proposed a centralized image interpretation system for urban rail security inspections. This system transmits X-ray images in slices to an image interpretation task scheduling server. After centralized processing by the image interpretation task scheduling server, the images are then distributed to the image interpretation terminals at a centralized image interpretation center through a scheduling algorithm. This centralized image interpretation center performs the image interpretation. This model theoretically could save 40% of the image interpretation staff through human resource reuse. However, in practice, the following issues remain: Image interpretation staff are highly demanding, typically requiring a break after every 15 minutes of interpretation. When break time is added, the theoretical savings in image interpretation staff are significantly reduced, significantly falling short of the expected savings. During actual security inspections, the centralized image interpretation system may become unavailable due to system or network failures at the image interpretation center, necessitating on-site image interpretation. However, on-site security inspectors are insufficient, and the remote location of image interpretation centers prevents timely replacement, resulting in insufficient security personnel. Summary of the Invention

[0005] The embodiments of the present application provide a decentralized image interpretation method and system for urban rail transit security inspection, which can solve the problem of insufficient human resources for image interpretation, improve the utilization rate of personnel in image interpretation work, and improve the efficiency of image interpretation work.

[0006] In a first aspect, an embodiment of the present application provides a method for decentralized image recognition in urban rail transit security inspection, comprising:

[0007] Receive in real time the scanned image of the inspected object collected by the security inspection equipment and the intelligent image recognition result of the scanned image by the intelligent image recognition equipment;

[0008] The received scanned image and the corresponding intelligent image recognition result are packaged to form an image recognition task, wherein the image recognition task carries the identification information of the security inspection equipment;

[0009] Assign the image judgment task to an idle image judgment station or re-inspection station, wherein each station is provided with at least one image judgment station and at least one security checkpoint, and each security checkpoint is provided with at least one security inspection device and at least one re-inspection station;

[0010] receiving the image judgment result fed back by the image judgment station or the re-inspection station, wherein the image judgment result carries the identification information of the security inspection equipment;

[0011] The image judgment result is sent to the re-inspection station of the security inspection equipment corresponding to the security inspection equipment identification information.

[0012] Furthermore, the re-inspection platform includes a re-inspection program and an image judgment program;

[0013] The method further comprises:

[0014] Use camera equipment to detect in real time whether there are passengers entering the preset range of the security inspection equipment;

[0015] If no passenger is detected entering the preset range of the security inspection device within the preset time, the re-inspection station corresponding to the security inspection device is determined to be idle, and the program in the re-inspection station is switched from the re-inspection program to the image judgment program, and the image judgment task is assigned to the re-inspection station for the re-inspector to perform the image judgment;

[0016] When the re-inspection station is in the image recognition process, it will detect in real time whether there are passengers entering the preset range of the security inspection equipment;

[0017] If it is detected that a passenger enters the preset range of the security inspection equipment, the image judgment task is stopped from being assigned to the re-inspection station, and the re-inspection station program is switched from the image judgment program to the re-inspection program after the current image judgment task is completed.

[0018] Furthermore, the image judgment task is assigned to an idle image judgment station, specifically:

[0019] The status information sent by the image judging station is received in real time, and the image judging task is assigned to the image judging station in an idle state according to the status information of the image judging station.

[0020] Furthermore, the security inspection equipment includes an X-ray baggage inspection machine, which is installed at the security inspection point and is used to collect X-ray scan images of baggage items in real time;

[0021] The real-time reception of the scanned image of the inspected object collected by the security inspection equipment and the intelligent image recognition result of the scanned image by the intelligent image recognition device is specifically:

[0022] Receive in real time the X-ray scan image slice data of the baggage items collected by the X-ray baggage inspection machine;

[0023] The fragmented data is sent to an intelligent image recognition device, which is used to merge the fragmented data into a complete image of the same luggage and perform intelligent image recognition to obtain an intelligent image recognition result, which carries the identification information of the security inspection device;

[0024] Receive the intelligent image recognition result fed back by the intelligent image recognition device.

[0025] Furthermore, the method further comprises:

[0026] Conduct fatigue detection and on-the-job detection on judges or reviewers through video equipment;

[0027] When fatigue is detected in the judge or reviewer, a fatigue reminder will be issued to prompt the judge or reviewer to take a rest;

[0028] When receiving the exit image judgment task instruction, the image judgment task is exited and no new image judgment task is assigned to the image judgment station or the review station until receiving the entry image judgment task instruction, and then the image judgment task is reallocated to the image judgment station or the review station.

[0029] Furthermore, the method further comprises:

[0030] Receive the switching program instruction sent by the re-inspection station;

[0031] The review program and the image judgment program are switched according to the switching program instruction.

[0032] Furthermore, after allocating the image judgment task to an idle image judgment station or a review station, the process also includes:

[0033] Start the image judgment countdown, which is used to remind the image judge or reviewer to feedback the image judgment result before the countdown ends;

[0034] If the image judgment result has not been received when the countdown is completed, the default image judgment result is that there are no dangerous items in the detected object.

[0035] In a second aspect, an embodiment of the present application provides a decentralized image recognition system for urban rail transit security inspection, comprising:

[0036] An image acquisition and intelligent recognition module is used to receive in real time the scanned image of the inspected object acquired by the security inspection equipment and the intelligent image recognition result of the scanned image by the intelligent image recognition equipment;

[0037] An image recognition task generation module is used to package the received scanned image and the corresponding intelligent image recognition result to form an image recognition task, wherein the image recognition task carries the identification information of the security inspection equipment;

[0038] An image judgment task allocation module is used to allocate the image judgment task to an idle image judgment station or re-inspection station, wherein each station is equipped with at least one image judgment station and at least one security checkpoint, and each security checkpoint is equipped with at least one security inspection device and at least one re-inspection station;

[0039] An image judgment result receiving module is used to receive the image judgment result fed back by the image judgment station or the re-inspection station, wherein the image judgment result carries the identification information of the security inspection equipment;

[0040] The image judgment result sending module is used to send the image judgment result to the re-inspection station of the security inspection equipment corresponding to the security inspection equipment identification information.

[0041] Furthermore, the re-inspection platform includes a re-inspection program and an image judgment program;

[0042] The system further includes a real-time monitoring module, which is used to detect in real time whether a passenger enters a preset range of the security inspection device through a camera device;

[0043] If no passenger is detected entering the preset range of the security inspection device within the preset time, the re-inspection station corresponding to the security inspection device is determined to be idle, and the program in the re-inspection station is switched from the re-inspection program to the image judgment program, and the image judgment task is assigned to the re-inspection station for the re-inspector to perform the image judgment;

[0044] When the re-inspection station is in the image recognition process, it will detect in real time whether there are passengers entering the preset range of the security inspection equipment;

[0045] If it is detected that a passenger enters the preset range of the security inspection equipment, the image judgment task is stopped from being assigned to the re-inspection station, and the re-inspection station program is switched from the image judgment program to the re-inspection program after the current image judgment task is completed.

[0046] Furthermore, the image judgment task allocation module is also used to receive status information sent by the image judgment station in real time, and allocate image judgment tasks to the image judgment station in an idle state according to the status information of the image judgment station.

[0047] Furthermore, the security inspection equipment includes an X-ray baggage inspection machine, which is installed at the security inspection point and is used to collect X-ray scan images of baggage items in real time;

[0048] The image acquisition and intelligent recognition module is also used to receive in real time the X-ray scan image slice data of the baggage items collected by the X-ray baggage inspection machine;

[0049] The fragmented data is sent to an intelligent image recognition device, which is used to merge the fragmented data into a complete image of the same luggage and perform intelligent image recognition to obtain an intelligent image recognition result, which carries the identification information of the security inspection device;

[0050] Receive the intelligent image recognition result fed back by the intelligent image recognition device.

[0051] Furthermore, the system further includes a personnel status detection module, which is used to perform fatigue detection and on-the-job detection on the image judge or reviewer through a camera device;

[0052] When fatigue is detected in the judge or reviewer, a fatigue reminder will be issued to prompt the judge or reviewer to take a rest;

[0053] When receiving the exit image judgment task instruction, the image judgment task is exited and no new image judgment task is assigned to the image judgment station or the review station until receiving the entry image judgment task instruction, and then the image judgment task is reallocated to the image judgment station or the review station.

[0054] Furthermore, the system also includes a manual switching module for receiving a switching program instruction sent by the re-inspection station;

[0055] The review program and the image judgment program are switched according to the switching program instruction.

[0056] Furthermore, the system further comprises a countdown module for starting a judgment countdown after the judgment task is assigned to an idle judgment station or review station, wherein the countdown is used to remind the judge or reviewer to provide feedback on the judgment result before the countdown is completed;

[0057] If the image judgment result has not been received when the countdown is completed, the default image judgment result is that there are no dangerous items in the detected object.

[0058] In a third aspect, an embodiment of the present application provides an electronic device, including:

[0059] memory and one or more processors;

[0060] The memory is used to store one or more programs;

[0061] When the one or more programs are executed by the one or more processors, the one or more processors implement the urban rail transit security inspection decentralized image judgment method as described in the first aspect.

[0062] In a fourth aspect, an embodiment of the present application provides a storage medium comprising computer-executable instructions, which, when executed by a computer processor, are used to execute the urban rail transit security inspection decentralized image judgment method as described in the first aspect.

[0063] The embodiment of the present application receives in real time the scanned image of the inspected object collected by the security inspection equipment and the intelligent image recognition result of the scanned image by the intelligent image recognition device, packages the received scanned image and the corresponding intelligent image recognition result into an image judgment task carrying the identification information of the security inspection equipment, and assigns the image judgment task to an idle image judgment station or re-inspection station, wherein each train station is provided with at least one image judgment station and at least one security inspection point, and each security inspection point is provided with at least one security inspection equipment and at least one re-inspection station, receives the image judgment result fed back by the image judgment station or re-inspection station, and sends the image judgment result to the re-inspection station of the security inspection equipment corresponding to the identification information. By adopting the above-mentioned technical means, by setting at least one image judgment station at each train station and sinking the image judgment station to the security inspection point, it is possible to rotate image judges with on-site security inspectors, making the rotation and deployment of image judges more flexible and improving the personnel utilization rate of image judgment work. In addition, the idle re-inspection desks are added to the image interpretation task allocation, so that when there are no passengers entering the security checkpoint, the re-inspectors can participate in the image interpretation work at the re-inspection desks, optimizing personnel utilization and improving the efficiency of image interpretation work. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 This is a flow chart of a decentralized image recognition method for urban rail transit security inspection provided in Example 1 of the present application;

[0065] Figure 2 This is a schematic diagram of device connection in Example 1 of the present application;

[0066] Figure 3 This is a structural diagram of a decentralized image recognition device for urban rail transit security inspection provided in Example 2 of the present application;

[0067] Figure 4 This is a structural diagram of an electronic device provided in Example 3 of the present application. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical solutions and advantages of the present application clearer, the specific embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. It is understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. It should also be noted that, for ease of description, only parts related to the present application, not all of the contents, are shown in the accompanying drawings. Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe each operation (or step) as a sequential process, many of the operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but can also have additional steps not included in the accompanying drawings. The process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0069] The urban rail transit security inspection decentralized image judgment method and system provided by the present application is intended to improve the utilization rate of personnel in the image judgment work and improve the efficiency of the image judgment work by allocating the image judgment task to the image judgment station or re-inspection station that is in an idle state at the station. Compared with the traditional centralized image judgment method, it usually sets up an image judgment center and concentrates the image judgment personnel in the image judgment center to perform image judgment work. Since the image judges need to take a break for a period of time after working for about 15 minutes before continuing to work, this results in a waste of part of the working time, resulting in insufficient work. In the actual security inspection process, the centralized image judgment system may be unavailable due to reasons such as system failure or network failure of the image judgment center, and the on-site image judgment mode needs to be adopted. The image judges in the image judgment center cannot be replenished to the security inspection point in time due to the long distance, resulting in insufficient on-site security inspection manpower. Based on this, the urban rail transit security inspection decentralized image judgment method of the embodiment of the present application is provided. It solves the problem of insufficient human resources for existing image judgment, improves the utilization rate of personnel in image judgment work, and improves the efficiency of image judgment work.

[0070] Example 1:

[0071] Figure 1 A flowchart of a decentralized image recognition method for urban rail transit security inspection provided in Example 1 of the present application is provided. The decentralized image recognition method for urban rail transit security inspection provided in this embodiment can be performed by a decentralized image recognition device for urban rail transit security inspection. The decentralized image recognition device for urban rail transit security inspection can be implemented through software and / or hardware. The decentralized image recognition device for urban rail transit security inspection can be composed of two or more physical entities or a single physical entity. Generally speaking, the decentralized image recognition device for urban rail transit security inspection can be a terminal device, such as a computer device.

[0072] The following description is based on the example of a computer device as the main body of the urban rail transit security inspection decentralized image recognition method. Figure 1 The urban rail transit security inspection decentralized image recognition method specifically includes:

[0073] S101: receiving in real time a scanned image of an inspected object collected by a security inspection device and an intelligent image recognition result of the scanned image by an intelligent image recognition device.

[0074] Security inspection equipment can include X-ray baggage inspection machines, security gates, or handheld metal detectors. Urban rail transit includes subways, buses, and high-speed trains. Security inspection equipment is installed at security checkpoints at each station, which are located at the station entrances. When passengers pass through the station security checkpoints, internet technology is used to intelligently collect, process, and distribute scanned images from multiple X-ray machines in the same or different locations. Multiple image interpretation terminals in the same or different locations then interpret the X-ray scans, recording and returning the interpretation results to the re-inspection station at the security checkpoint for display.

[0075] Reference Figure 2 Each train station is equipped with at least one image judgment station and at least one security checkpoint, and each security checkpoint is equipped with at least one security inspection device and at least one re-inspection station. When a station has multiple security checkpoints, the number of image judgment stations is set according to the ratio. For example, if the ratio of the total number of security checkpoints to the expected maximum number of image judges is 5:3, then a station with 5 security checkpoints can set up 3 image judgment stations to achieve the purpose of evenly distributing image judges, and try to configure image judgment stations at security checkpoints with relatively large passenger flow. The specific ratio of the number of security checkpoints to the number of image judgment stations can be set according to actual conditions and is not limited in this embodiment. All image judgment stations are interconnected with all security checkpoints and image judgment task scheduling servers through the security inspection network. All re-inspection stations are also interconnected with all security checkpoints and image judgment task scheduling servers.

[0076] The security inspection equipment includes an X-ray baggage inspection machine, installed at a security checkpoint, for collecting X-ray scanned images of baggage items in real time. The X-ray machine scans the baggage items to obtain multiple columns of line scan data. The obtained columns of line scan data are colorized to produce a pseudo-color X-ray scanned image with material information. The X-ray scanned image is a digital image. Sliced ​​data of the X-ray scanned image is sent to an intelligent image recognition device, which merges the sliced ​​data to form a complete image of the baggage and performs intelligent image recognition to obtain an intelligent image recognition result. Suspected prohibited items are marked on the intelligent image recognition result to alert the image judge. The intelligent image recognition device can correspond one-to-one with each security inspection device, or one intelligent image recognition device can perform intelligent image recognition on multiple security inspection devices. The sliced ​​data carries the identification information of the corresponding security inspection device performing the X-ray scan, resulting in the intelligent image recognition result carrying the identification information of the security inspection device. The X-ray scanned image sliced ​​data of the baggage items collected by the X-ray baggage inspection machine and the intelligent image recognition result fed back by the intelligent image recognition device are received in real time to prepare for the subsequent image recognition task.

[0077] S102: Pack the received scanned image and the corresponding intelligent image recognition result to form an image recognition task, wherein the image recognition task carries the identification information of the security inspection equipment.

[0078] For the received scanned image slice data sent by the security inspection device and the security inspection device identification information carried in the slice data, and the intelligent image recognition result sent by the intelligent image recognition device and the security inspection device identification information carried in the intelligent image recognition result, the scanned image slice data and the intelligent image recognition result data corresponding to the same security inspection device identification information are packaged to form an image judgment task. The image judgment task carries the security inspection device identification information, so that after the image judgment is completed, the image judgment result is fed back to the re-inspection station corresponding to the corresponding security inspection device according to the security inspection device identification information.

[0079] S103, allocating the image judgment task to an idle image judgment station or re-inspection station, wherein each station is provided with at least one image judgment station and at least one security checkpoint, and each security checkpoint is provided with at least one security inspection device and at least one re-inspection station.

[0080] The system receives status information from image recognition stations in real time and assigns image recognition tasks to idle image recognition stations based on the status information of the image recognition stations. The status information of the image recognition stations includes two states: assignable and unassignable. The assignable state is an idle state, and image recognition tasks can only be assigned to image recognition stations in the assignable state.

[0081] The re-inspection station at the security checkpoint includes a re-inspection program and an image judgment program. A camera is used to detect in real time whether a passenger has entered the preset range of the security checkpoint. If no passenger is detected within the preset range within a preset time, the re-inspection station corresponding to the security checkpoint is determined to be idle, and the program in the re-inspection station is switched from the re-inspection program to the image judgment program. The image judgment task is then assigned to the re-inspection station for the re-inspector to perform image judgment. While the re-inspection station is in the image judgment program, it detects in real time whether a passenger has entered the preset range of the security checkpoint. If a passenger is detected entering the preset range of the security checkpoint, the assignment of the image judgment task to the re-inspection station is stopped, and after the current image judgment task is completed, the re-inspection station program is switched from the image judgment program to the re-inspection program. There is a certain time difference between detecting that a passenger enters the preset range of the security inspection equipment and the time when the passenger passes the security inspection, which is enough for the re-inspector to complete the current image judgment task. Therefore, when it is detected that a passenger enters the preset range of the security inspection equipment, a feedback message of stopping the allocation of image judgment tasks is sent to the image judgment task scheduling server to stop allocating image judgment tasks to the re-inspection desk, and after the re-inspector completes the current image judgment task, the re-inspection desk program is switched from the image judgment program to the re-inspection program to carry out subsequent re-inspection work.

[0082] For example, a preset timer of five minutes is used. A camera monitors the security checkpoint's security equipment in real time to detect passengers entering the preset area for security checks. If no passengers are detected within the preset area within five minutes, the current time period is considered off-peak, indicating a low number of passengers entering the station for security checks. Therefore, both the re-inspector and the re-inspection desk are idle. Therefore, the re-inspection desk's program is switched from the re-inspection program to the image interpretation program, and the image interpretation task is assigned to the re-inspection desk for the re-inspector to perform image interpretation. This significantly improves the rationality of the re-inspector's work schedule during idle time periods and improves resource utilization at the security checkpoint. While the re-inspector is performing image interpretation, a passenger may suddenly enter the station. Therefore, real-time detection is required within the preset area of ​​the security checkpoint equipment. When a passenger is detected entering, the re-inspection desk stops assigning image interpretation tasks to the corresponding re-inspection desk. After the current image interpretation task is completed, the re-inspection desk's program is switched from the image interpretation program to the re-inspection program. In this embodiment, the preset time is set to 5 minutes for illustration only. The specific preset time setting can be set according to actual conditions. For example, at a station in a remote suburb, the passenger flow entering the station is relatively small throughout the day, and the preset time can be set to a smaller time, such as 1 minute.

[0083] In one embodiment, a program switching instruction sent by the re-inspection station is received, and the re-inspection program and the image judgment program are switched according to the program switching instruction. In this embodiment, the program mode can be manually switched by the re-inspector, and the re-inspection station receives the program switching instruction input by the re-inspector and sends the program switching instruction to the image judgment task scheduling server. The image judgment task scheduling server receives the program switching instruction sent by the re-inspection station and switches the re-inspection program and the image judgment program according to the program switching instruction. When there are no passengers entering the station for security inspection for a period of time, but the preset automatic switching time has not yet arrived, the program of the re-inspection station can be switched from the re-inspection program to the image judgment program by manual switching, and the program switching instruction is sent to the image judgment task scheduling server. The image judgment task scheduling server receives the program switching instruction sent by the re-inspection station, switches the re-inspection program to the image judgment program according to the program switching instruction, and assigns the image judgment task to the corresponding re-inspection station for the re-inspector to perform image judgment work. When the re-inspection station is in the image judgment program, due to an emergency, the re-inspector needs to suspend the image judgment work to deal with the emergency. The program of the re-inspection station can be switched from the image judgment program to the re-inspection program by manual switching, and the program switching instruction is sent to the image judgment task scheduling server. The image judgment task server receives the program switching instruction sent by the re-inspection station, stops allocating image judgment tasks to the corresponding re-inspection station, and switches the program in the re-inspection station from the image judgment program to the re-inspection program after completing the current image judgment task.

[0084] S104: Receive the image judgment result fed back by the image judgment station or the re-inspection station, wherein the image judgment result carries the identification information of the security inspection equipment.

[0085] Receive the image judgment result information fed back by the image judgment station or re-inspection station. If the image judgment result is that there are no dangerous goods, directly feedback the security check pass result information. If the image judgment result is that there are dangerous goods, feedback the security check fail result information and feedback the scanned image marked with suspected dangerous goods.

[0086] In one embodiment, after the image judgment task is assigned to an idle image judgment station or re-inspection station, a judgment countdown is initiated. This is because if dangerous goods are found in the luggage, the re-inspector at the re-inspection station must be notified to conduct the corresponding unpacking and confirmation inspection. Therefore, a judgment countdown is required, and the judgment result must be output within a preset time. The countdown is used to remind the image judge or re-inspector to feedback the judgment result before the countdown expires. If no judgment result feedback is received by the time the countdown expires, the default judgment result is that the inspected object does not contain any dangerous goods and the inspected object can pass security inspection.

[0087] S105: Send the image recognition result to the re-inspection station of the security inspection equipment corresponding to the security inspection equipment identification information.

[0088] If the image judgment result indicates that no dangerous goods are present, the security inspection pass result information will be directly fed back to the re-inspection station corresponding to the security inspection equipment corresponding to the equipment identification information of the image judgment task, so as to remind the re-inspector to release the luggage. If the image judgment result indicates that dangerous goods are present, the security inspection fail result information will be fed back, and the scanned image marked with suspected dangerous goods will be fed back to the re-inspection station corresponding to the equipment identification information of the image judgment task, so as to remind the re-inspector to intercept the corresponding luggage item and conduct an unpacking inspection to confirm the presence of dangerous goods.

[0089] In one embodiment, fatigue detection and on-duty detection are performed on the judge or reviewer using a camera. When fatigue is detected in the judge or reviewer, a fatigue reminder is issued to prompt the judge or reviewer to take a break. When an exit judgment task instruction is received, the judge task is exited and no new judgment tasks are assigned to the judge station or review desk until an entry judgment task instruction is received, at which point the judge task is reassigned to the judge station or review desk. If the judge inputs an exit judgment task instruction at the judge station or the reviewer inputs an exit judgment task instruction at the review desk, the judge or reviewer is deemed to be off-duty, the judge task is exited, and no new judgment tasks are assigned to the judge station or review desk. If an entry judgment task instruction is received, the judge or reviewer is deemed to be on-duty, and the judge task is assigned to the corresponding judge station or review desk for the on-duty judge or reviewer to perform the corresponding judgment work.

[0090] In the above, by receiving in real time the scanned image of the inspected object collected by the security inspection equipment and the intelligent image recognition result of the scanned image by the intelligent image recognition device, the received scanned image and the corresponding intelligent image recognition result are packaged to form an image judgment task carrying the identification information of the security inspection equipment, and the image judgment task is assigned to an idle image judgment station or re-inspection station, wherein each train station is equipped with at least one image judgment station and at least one security inspection point, and each security inspection point is equipped with at least one security inspection equipment and at least one re-inspection station, receiving the image judgment result fed back by the image judgment station or re-inspection station, and sending the image judgment result to the re-inspection station of the security inspection equipment corresponding to the identification information. By adopting the above technical means, by setting at least one image judgment station at each train station and sinking the image judgment station to the security inspection point, it is possible to rotate image judges with on-site security inspectors, making the rotation and deployment of image judges more flexible and improving the personnel utilization rate of image judgment work. In addition, the idle re-inspection desks are added to the image interpretation task allocation, so that when there are no passengers entering the security checkpoint, the re-inspectors can participate in the image interpretation work at the re-inspection desks, optimizing personnel utilization and improving the efficiency of image interpretation work.

[0091] Example 2:

[0092] Based on the above embodiments, Figure 3This is a schematic diagram of the structure of a distributed image recognition device for urban rail transit security inspection provided in Example 2 of this application. Figure 3 The urban rail transit security inspection decentralized image recognition device provided in this embodiment specifically includes: an image acquisition and intelligent recognition module 21, an image recognition task generation module 22, an image recognition task allocation module 23, an image recognition result receiving module 24 and an image recognition result sending module 25.

[0093] The image acquisition and intelligent recognition module 21 is used to receive in real time the scanned image of the inspected object acquired by the security inspection equipment and the intelligent recognition result of the scanned image by the intelligent image recognition equipment;

[0094] An image recognition task generating module 22 is configured to package the received scanned image and the corresponding intelligent image recognition result to form an image recognition task, wherein the image recognition task carries the identification information of the security inspection equipment;

[0095] The image judgment task allocation module 23 is used to allocate the image judgment task to an idle image judgment station or re-inspection station, wherein each station is equipped with at least one image judgment station and at least one security checkpoint, and each security checkpoint is equipped with at least one security inspection device and at least one re-inspection station;

[0096] The image judgment result receiving module 24 is used to receive the image judgment result fed back by the image judgment station or the re-inspection station, wherein the image judgment result carries the identification information of the security inspection equipment;

[0097] The image recognition result sending module 25 is used to send the image recognition result to the re-inspection station of the security inspection equipment corresponding to the security inspection equipment identification information.

[0098] Furthermore, the re-inspection platform includes a re-inspection program and an image judgment program;

[0099] The system further includes a real-time monitoring module, which is used to detect in real time whether a passenger enters a preset range of the security inspection device through a camera device;

[0100] If no passenger is detected entering the preset range of the security inspection device within the preset time, the re-inspection station corresponding to the security inspection device is determined to be idle, and the program in the re-inspection station is switched from the re-inspection program to the image judgment program, and the image judgment task is assigned to the re-inspection station for the re-inspector to perform the image judgment;

[0101] When the re-inspection station is in the image recognition process, it will detect in real time whether there are passengers entering the preset range of the security inspection equipment;

[0102] If it is detected that a passenger enters the preset range of the security inspection equipment, the image judgment task is stopped from being assigned to the re-inspection station, and the re-inspection station program is switched from the image judgment program to the re-inspection program after the current image judgment task is completed.

[0103] Furthermore, the image judgment task assignment module 23 is further configured to receive status information sent by the image judgment stations in real time, and assign image judgment tasks to idle image judgment stations according to the status information of the image judgment stations.

[0104] Furthermore, the security inspection equipment includes an X-ray baggage inspection machine, which is installed at the security inspection point and is used to collect X-ray scan images of baggage items in real time;

[0105] The image acquisition and intelligent recognition module 21 is also used to receive in real time the X-ray scan image slice data of the baggage items collected by the X-ray baggage inspection machine;

[0106] The fragmented data is sent to an intelligent image recognition device, which is used to merge the fragmented data into a complete image of the same luggage and perform intelligent image recognition to obtain an intelligent image recognition result, which carries the identification information of the security inspection device;

[0107] Receive the intelligent image recognition result fed back by the intelligent image recognition device.

[0108] Furthermore, the system further includes a personnel status detection module, which is used to perform fatigue detection and on-the-job detection on the image judge or reviewer through a camera device;

[0109] When fatigue is detected in the judge or reviewer, a fatigue reminder will be issued to prompt the judge or reviewer to take a rest;

[0110] When receiving the exit image judgment task instruction, the image judgment task is exited and no new image judgment task is assigned to the image judgment station or the review station until receiving the entry image judgment task instruction, and then the image judgment task is reallocated to the image judgment station or the review station.

[0111] Furthermore, the system also includes a manual switching module for receiving a switching program instruction sent by the re-inspection station;

[0112] The review program and the image judgment program are switched according to the switching program instruction.

[0113] Furthermore, the system further comprises a countdown module for starting a judgment countdown after the judgment task is assigned to an idle judgment station or review station, wherein the countdown is used to remind the judge or reviewer to provide feedback on the judgment result before the countdown is completed;

[0114] If the image judgment result has not been received when the countdown is completed, the default image judgment result is that there are no dangerous items in the detected object.

[0115] In the above, by receiving in real time the scanned image of the inspected object collected by the security inspection equipment and the intelligent image recognition result of the scanned image by the intelligent image recognition device, the received scanned image and the corresponding intelligent image recognition result are packaged to form an image judgment task carrying the identification information of the security inspection equipment, and the image judgment task is assigned to an idle image judgment station or re-inspection station, wherein each train station is equipped with at least one image judgment station and at least one security inspection point, and each security inspection point is equipped with at least one security inspection equipment and at least one re-inspection station, receiving the image judgment result fed back by the image judgment station or re-inspection station, and sending the image judgment result to the re-inspection station of the security inspection equipment corresponding to the identification information. By adopting the above technical means, by setting at least one image judgment station at each train station and sinking the image judgment station to the security inspection point, it is possible to rotate image judges with on-site security inspectors, making the rotation and deployment of image judges more flexible and improving the personnel utilization rate of image judgment work. In addition, by adding idle re-inspection stations to the image judgment task allocation, when there are no passengers entering the station for security inspection, re-inspectors can participate in the image judgment work at the re-inspection stations, optimizing personnel utilization and improving image judgment efficiency. The urban rail transit security inspection decentralized image judgment device provided in Example 2 of this application can be used to execute the urban rail transit security inspection decentralized image judgment method provided in Example 1 above, and has corresponding functions and beneficial effects.

[0116] Example 3:

[0117] The third embodiment of the present application provides an electronic device, referring to Figure 4 The electronic device includes: a processor 31, a memory 32, a communication module 33, an input device 34, and an output device 35. The number of processors in the electronic device may be one or more, and the number of memories in the electronic device may be one or more. The processor, memory, communication module, input device, and output device of the electronic device may be connected via a bus or other means.

[0118] The memory 32 is a computer-readable storage medium that can be used to store software programs, computer executable programs, and modules, such as the program instructions / modules corresponding to the urban rail transit security inspection decentralized image discrimination method described in any embodiment of the present application (for example, the image acquisition and intelligent recognition module, the image discrimination task generation module, the image discrimination task allocation module, the image discrimination result receiving module, and the image discrimination result sending module in the urban rail transit security inspection decentralized image discrimination device). The memory may mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system and at least one application required for a function; the data storage area can store data created according to the use of the device, etc. In addition, the memory may include a high-speed random access memory and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory may further include a memory remotely located relative to the processor, and these remote memories can be connected to the device via a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0119] The communication module 33 is used for data transmission.

[0120] The processor 31 executes various functional applications and data processing of the device by running the software programs, instructions and modules stored in the memory, that is, realizes the above-mentioned urban rail transit security inspection decentralized image judgment method.

[0121] The input device 34 may be used to receive input digital or character information and generate key signal input related to user settings and function control of the device. The output device 35 may include a display device such as a display screen.

[0122] The electronic device provided above can be used to execute the urban rail transit security inspection decentralized image recognition method provided in the above embodiment 1, and has corresponding functions and beneficial effects.

[0123] Example 4:

[0124] An embodiment of the present application also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to execute a decentralized image judgment method for urban rail transit security inspection, the decentralized image judgment method for urban rail transit security inspection comprising: receiving in real time a scanned image of an inspected object collected by a security inspection device and an intelligent image recognition result of the scanned image by an intelligent image recognition device; packaging the received scanned image and the corresponding intelligent image recognition result to form an image judgment task, the image judgment task carrying security inspection device identification information; allocating the image judgment task to an idle image judgment station or re-inspection station, wherein each train station is provided with at least one image judgment station and at least one security inspection point, and each security inspection point is provided with at least one security inspection device and at least one re-inspection station; receiving an image judgment result fed back by the image judgment station or re-inspection station, the image judgment result carrying the security inspection device identification information; and sending the image judgment result to the re-inspection station of the security inspection device corresponding to the security inspection device identification information.

[0125] Storage medium - any of various types of memory devices or storage devices. The term "storage medium" is intended to include: installation media, such as CD-ROMs, floppy disks, or tape drives; computer system memory or random access memory, such as DRAM, DDR RAM, SRAM, EDO RAM, Rambus RAM, etc.; non-volatile memory, such as flash memory, magnetic media (such as hard disks or optical storage); registers or other similar types of memory elements, etc. Storage media may also include other types of memory or combinations thereof. In addition, the storage medium may be located in the first computer system in which the program is executed, or it may be located in a different second computer system that is connected to the first computer system via a network (such as the Internet). The second computer system can provide program instructions to the first computer for execution. The term "storage medium" may include two or more storage media residing in different locations (e.g., in different computer systems connected via a network). The storage medium may store program instructions (e.g., embodied as a computer program) that can be executed by one or more processors.

[0126] Of course, the storage medium containing computer-executable instructions provided in an embodiment of the present application, whose computer-executable instructions are not limited to the urban rail transit security inspection decentralized image judgment method described above, can also execute related operations in the urban rail transit security inspection decentralized image judgment method provided in any embodiment of the present application.

[0127] The urban rail transit security inspection decentralized image judgment device, storage medium and electronic device provided in the above embodiments can execute the urban rail transit security inspection decentralized image judgment method provided in any embodiment of the present application. For technical details not described in detail in the above embodiments, please refer to the urban rail transit security inspection decentralized image judgment method provided in any embodiment of the present application.

[0128] The above are only preferred embodiments of the present application and the technical principles employed. The present application is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions that are possible for those skilled in the art will not depart from the scope of protection of the present application. Therefore, although the present application has been described in more detail through the above embodiments, the present application is not limited to the above embodiments and may include more other equivalent embodiments without departing from the concept of the present application. The scope of the present application is determined by the scope of the claims.

Claims

1. A decentralized image recognition method for urban rail transit security inspection, characterized in that: include: Receive in real time the scanned image of the inspected object collected by the security inspection equipment and the intelligent image recognition result of the scanned image by the intelligent image recognition equipment; The received scanned image and the corresponding intelligent image recognition result are packaged to form an image recognition task, wherein the image recognition task carries the identification information of the security inspection equipment; Assign the image judgment task to an idle image judgment station or re-inspection station, wherein each station is equipped with at least one image judgment station and at least one security checkpoint, and each security checkpoint is equipped with at least one security inspection device and at least one re-inspection station, wherein the re-inspection station includes a re-inspection program and an image judgment program; receiving the image judgment result fed back by the image judgment station or the re-inspection station, wherein the image judgment result carries the identification information of the security inspection equipment; Sending the image recognition result to the re-inspection station of the security inspection device corresponding to the identification information of the security inspection device; The method further comprises: Use camera equipment to detect in real time whether there are passengers entering the preset range of the security inspection equipment; If no passenger is detected entering the preset range of the security inspection equipment within the preset time, it is determined that the re-inspection table corresponding to the security inspection equipment is in an idle state, and the program in the re-inspection table is switched from the re-inspection program to the image judgment program, and the image judgment task is assigned to the re-inspection table for the re-inspector to perform image judgment.

2. The urban rail transit security inspection decentralized image recognition method according to claim 1, characterized in that: The method further comprises: When the re-inspection station is in the image recognition process, it will detect in real time whether there are passengers entering the preset range of the security inspection equipment; If it is detected that a passenger enters the preset range of the security inspection equipment, the image judgment task is stopped from being assigned to the re-inspection station, and the re-inspection station program is switched from the image judgment program to the re-inspection program after the current image judgment task is completed.

3. The urban rail transit security inspection decentralized image recognition method according to claim 1, characterized in that: The step of allocating the image judgment task to an idle image judgment station is as follows: The status information sent by the image judging station is received in real time, and the image judging task is assigned to the image judging station in an idle state according to the status information of the image judging station.

4. The urban rail transit security inspection decentralized image recognition method according to claim 1, characterized in that: The security inspection equipment includes an X-ray baggage inspection machine, which is set up at the security inspection point and is used to collect X-ray scan images of baggage items in real time; The real-time reception of the scanned image of the inspected object collected by the security inspection equipment and the intelligent image recognition result of the scanned image by the intelligent image recognition device is specifically: Receive in real time the X-ray scan image slice data of the baggage items collected by the X-ray baggage inspection machine; The fragmented data is sent to an intelligent image recognition device, which is used to merge the fragmented data into a complete image of the same luggage and perform intelligent image recognition to obtain an intelligent image recognition result, which carries the identification information of the security inspection device; Receive the intelligent image recognition result fed back by the intelligent image recognition device.

5. The urban rail transit security inspection decentralized image recognition method according to claim 1, characterized in that: The method further comprises: Conduct fatigue detection and on-the-job detection on judges or reviewers through video equipment; When fatigue is detected in the judge or reviewer, a fatigue reminder will be issued to prompt the judge or reviewer to take a rest; When receiving the exit image judgment task instruction, the image judgment task is exited and no new image judgment task is assigned to the image judgment station or the review station until receiving the entry image judgment task instruction, and then the image judgment task is reallocated to the image judgment station or the review station.

6. The urban rail transit security inspection decentralized image recognition method according to claim 2, characterized in that: The method further comprises: Receive the switching program instruction sent by the re-inspection station; The review program and the image judgment program are switched according to the switching program instruction.

7. The urban rail transit security inspection decentralized image recognition method according to claim 1, characterized in that: After allocating the image judgment task to an idle image judgment station or review station, the following steps are also included: Start the image judgment countdown, which is used to remind the image judge or reviewer to feedback the image judgment result before the countdown ends; If the image judgment result has not been received when the countdown is completed, the default image judgment result is that there are no dangerous items in the detected object.

8. A decentralized image recognition system for urban rail transit security inspection, characterized in that: include: An image acquisition and intelligent recognition module is used to receive in real time the scanned image of the inspected object acquired by the security inspection equipment and the intelligent image recognition result of the scanned image by the intelligent image recognition equipment; An image recognition task generation module is used to package the received scanned image and the corresponding intelligent image recognition result to form an image recognition task, wherein the image recognition task carries the security inspection equipment identification information; An image judgment task allocation module is used to allocate the image judgment task to an idle image judgment station or re-inspection station, wherein each station is equipped with at least one image judgment station and at least one security checkpoint, and each security checkpoint is equipped with at least one security inspection device and at least one re-inspection station, and the re-inspection station includes a re-inspection program and an image judgment program; An image judgment result receiving module is used to receive the image judgment result fed back by the image judgment station or the re-inspection station, wherein the image judgment result carries the identification information of the security inspection equipment; An image judgment result sending module, configured to send the image judgment result to the re-inspection station of the security inspection equipment corresponding to the identification information of the security inspection equipment; The urban rail transit security inspection decentralized image recognition system also includes a real-time monitoring module; The real-time monitoring module is used to detect in real time whether there are passengers entering the preset range of the security inspection equipment through the camera equipment; If no passenger is detected entering the preset range of the security inspection equipment within the preset time, it is determined that the re-inspection table corresponding to the security inspection equipment is in an idle state, and the program in the re-inspection table is switched from the re-inspection program to the image judgment program, and the image judgment task is assigned to the re-inspection table for the re-inspector to perform image judgment.

9. An electronic device, characterized in that: include: memory and one or more processors; The memory is used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A storage medium containing computer-executable instructions, characterized in that: When the computer executable instructions are executed by a computer processor, they are used to perform the method according to any one of claims 1 to 7.

Images