Platform foreign matter identification method and system, electronic equipment and storage medium

Abstract

The invention relates to a platform foreign matter identification method and system, electronic equipment and a storage medium, and belongs to the technical field of image processing.The method comprises the steps that image information in a platform is collected; performing region division on the image information to obtain a to-be-recognized region and a non-recognized region of the image information; and identifying the to-be-identified area based on a preset foreign matter identification model to obtain a platform foreign matter identification result. According to the invention, the to-be-recognized area is recognized by using the preset foreign matter recognition model, so that the foreign matter recognition efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular, to a method, device, electronic device and storage medium for identifying foreign objects on a platform. Background Art

[0002] Before a train travels, people or foreign objects within the safety yellow line pose a great safety hazard to the train's travel. When a train enters or exits a station, someone must be on the platform of the station to observe and remind passengers in real time not to stand within the safety yellow line. However, it is inevitable to be negligent or have the vision blocked, resulting in safety hazards. In addition, small-sized objects are too far away, and it is difficult to see clearly with the naked eye.

[0003] Some teams in the market have studied using lasers for detection. The principle of the laser is to emit and receive at the same time. When the signal cannot be received, it can be determined that there is an obstacle or foreign object. However, the laser is greatly affected by the environment. It often gives false alarms on rainy days, and the installation stiffness requirements of the laser are relatively high. When the train enters or exits, the vibration shifts the laser emission or reception position, which will also cause false alarms. There are also image recognition-based methods and neural network-based pattern recognition methods in the market, but the recognition effects are not good.

[0004] Therefore, how to design a set of foreign object recognition methods to reduce the influence of the environment is an urgent technical problem to be solved. Summary of the Invention

[0005] In view of this, it is necessary to provide a method, system, electronic device and storage medium for identifying foreign objects in the safety area of a platform, which can reduce the influence of the environment when identifying foreign objects in the safety area of the platform.

[0006] To solve the above problems, in a first aspect, the present invention provides a method for identifying foreign objects on a platform, including: Collecting image information within the platform; Dividing the image information into regions to obtain a region to be recognized and a non-recognized region of the image information; Identifying the region to be recognized based on a preset foreign object recognition model to obtain a foreign object recognition result on the platform.

[0007] In a possible implementation manner, it further includes: Responding to the train departure signal and collecting image information within the platform.

[0008] In a possible implementation manner, the dividing the image information into regions to obtain a region to be recognized and a non-recognized region of the image information includes: Performing grayscale processing on the image information to obtain the grayscale value data corresponding to the image; Divide the image information into regions based on a preset grayscale value range to obtain the region to be recognized and the non-recognized region of the image information.

[0009] In a possible implementation manner, the generation process of the preset foreign object recognition model includes: Collect historical image information under the platform background, and label the historical image information based on preset foreign objects to obtain a labeled image information set, where the preset foreign objects are small items that are easily overlooked; Use a transfer learning model to learn the labeled image information set to generate a foreign object recognition model.

[0010] In a possible implementation manner, it further includes: If the platform foreign object recognition result is that no platform foreign object is recognized in the region to be recognized, identify the camera information of the next frame of image in the platform until all cameras have completed the round-robin.

[0011] In a possible implementation manner, it further includes: If the platform foreign object recognition result is that a platform foreign object is recognized in the region to be recognized, send an alarm signal to the train, and after the danger is lifted, send a safety signal to the train to enable the train to move forward.

[0012] In a possible implementation manner, the above method further includes: If the platform foreign object recognition result is that a platform foreign object is recognized in the region to be recognized; Statistically analyze the foreign object type information and foreign object location information.

[0013] In a second aspect, the present invention further provides a foreign object recognition system for the platform safety area. In a possible implementation manner, it includes: A camera, configured to collect image information in the platform in real time and transmit the image information to a switch through a network cable; A display and control room, configured to transmit the train approaching signal to the switch through an optical fiber; A switch, configured to transmit an identification signal and the image information to a server after receiving the train approaching signal; A server, configured to perform foreign object recognition on the image information through a preset foreign object recognition model after receiving the identification signal; A display, configured to display the foreign object recognition result.

[0014] In a possible implementation manner, the above system further includes: If the platform foreign object recognition result is that no platform foreign object is recognized in the region to be recognized, the server recognizes the next frame of image information until the display and control room sends a signal indicating no need for detection; If the platform foreign object recognition result is that a platform foreign object is recognized in the area to be recognized, when there is a foreign object, the server issues an alarm signal and a train stop signal, and transmits the alarm signal and the train stop signal to the switch through a network cable, so that the switch transmits the alarm signal and the train stop signal to the display and control room through an optical fiber.

[0015] In a third aspect, the present invention further provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned platform foreign object recognition method are implemented.

[0016] In a fourth aspect, the present invention further provides a computer storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned platform foreign object recognition method are implemented.

[0017] The beneficial effects of the present invention are as follows: By collecting image information inside the platform through a camera, the present invention improves the collection efficiency compared with manual work. And by dividing the area of the image inside the platform, the recognition area is reduced, the calculation amount in the recognition process is reduced, and by recognizing foreign objects inside the platform through images, the influence of the environment can be reduced. Finally, the preset foreign object recognition model is used to recognize the area to be recognized, improving the recognition efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those skilled in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0019] Figure 1 It is a flowchart of a method according to an embodiment of a platform foreign object recognition method provided by the present invention; Figure 2 It is a schematic structural diagram of a platform foreign object recognition system according to an embodiment provided by the present invention; Figure 3 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] The following will specifically describe the preferred embodiments of the present invention with reference to the drawings. Among them, the drawings constitute a part of this application and are used together with the embodiments of the present invention to explain the principle of the present invention, and are not used to limit the scope of the present invention.

[0021] References to "embodiments" in this document mean that the specific features, structures, or characteristics described in connection with the embodiments can be included in at least one embodiment of the present invention. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0022] A specific embodiment of the present invention, such as Figure 1 shown Figure 1 is a flowchart of a method for an embodiment of a method for identifying foreign objects on a platform provided by the present invention, including: Step 101: Collect image information within the platform; Step 102: Divide the image information into regions to obtain the regions to be recognized and non-recognized regions of the image information; Step 103: Based on a preset foreign object recognition model, recognize the regions to be recognized to obtain the foreign object recognition result on the platform.

[0023] The present invention collects image information within the platform through a camera, which improves the collection efficiency compared to manual work. And by dividing the regions of the images within the platform, the recognition region is reduced, the amount of calculation during the recognition process is decreased, and by recognizing foreign objects within the platform through images, the influence of laser detection vibration and rainy days can be reduced. Finally, a preset foreign object recognition model is used to recognize the regions to be recognized, improving the recognition efficiency.

[0024] In an embodiment of the present invention, dividing the image information into regions to obtain the regions to be recognized and non-recognized regions of the image information includes: Perform gray-scale processing on the image information to obtain the gray-scale value data corresponding to the image; Based on a preset gray-scale value range, divide the image information into regions to obtain the regions to be recognized and non-recognized regions of the image information.

[0025] It can be understood that the image information within the platform is collected in response to the train approaching signal. However, image comparison is greatly affected by light and vibration. The gray-scale value of the blank part in the image is relatively small, and the gray-scale value of the part with an object is relatively large. Therefore, gray-scale processing can be performed on the image information to further enhance the contrast. Specifically, the gray-scale value of the image information can be determined first, and then based on a preset gray-scale value range, for example, the region with a gray-scale value of 0 to 50 is the non-recognized region, and the region with a gray-scale value of 51 to 225 is the region to be recognized. By performing gray-scale processing on the image information, on the one hand, the influence of light and vibration can be reduced, and on the other hand, the regions to be recognized and non-recognized regions of the image information can be distinguished.

[0026] In one embodiment of the present invention, the generation process of the preset foreign object recognition model includes: Collect historical image information under the platform background, and label the historical image information based on the preset foreign objects to obtain a labeled image information set, where the preset foreign objects are small items that are easily overlooked; Use the transfer learning model to learn the labeled image information set to generate a foreign object recognition model.

[0027] First of all, it should be noted that after obtaining the labeled image information set, median filtering and multi-scale weighted average Retinex algorithm can also be used to denoise and enhance the labeled image information set to increase the recognition rate of the area to be recognized. In the prior art, the recognition mode based on neural network will miss the detection of unlearned models, while transfer learning is to use the knowledge learned from one environment to help the learning task in the new environment, reducing the risk of missed detection. Before the server performs recognition, it needs to learn and store the features in the server, and when processing image information, it will perform processing.

[0028] Specifically, collect various image information based on the platform background, label the foreign objects that are often easily missed; perform gray-scale processing on the images; use the transfer learning model to learn the data after gray-scale processing of the labels and store the learning results in the server; divide the recognition area of the collected data and remove the unnecessary areas; perform gray-scale processing on the divided recognition area and enhance the contrast of the recognition area to generate a foreign object recognition model. Finally, use the learned learning parameters to recognize the collected images. This method has good accuracy and can also classify and recognize videos with poor lighting, so it can have better practicability and recognition efficiency, and complete the recognition of one frame of the camera image.

[0029] Through the foreign object recognition model, it is possible to limit the influence of different lighting on the recognition of objects at different time periods. After the image is gray-scale processed, it can effectively reduce the influence of other areas of the image on the gray-scale processing effect. The platform and foreign objects will reduce the influence of light on the image, thus effectively avoiding the influence of sunlight at different time periods on the recognition effect. And because it is to recognize the features in the image, it can avoid the influence of vibration on the recognition effect. Finally, through the use of transfer learning, the problem of missed detection caused by insufficient data volume is compensated.

[0030] In one embodiment of the present invention, the above method further includes: If the platform foreign object recognition result is that no platform foreign object is recognized in the area to be recognized, identify the camera information of the next frame of the image in the platform until all cameras have completed the round-robin.

[0031] If the platform foreign object recognition result is that a platform foreign object is recognized in the area to be recognized, an alarm signal is sent to the train, and after the danger is lifted, a safety signal is sent to the train to enable the train to move forward.

[0032] It can be understood that during the time when the train enters the station to leaves the station, the camera collects the image information inside the platform in real time, that is, from the train entering the station to the train leaving the station, the camera collects multiple frames of images. If no foreign object is recognized in the currently collected image information, the recognition can continue for the next frame of camera image information until all camera polling is completed; if a foreign object is recognized in the currently collected image information summary, the foreign object alarm program is directly started to send an alarm signal to the train to ensure the safety during the train's travel, and after the danger is lifted, a safety signal is sent to the train to enable the train to move forward.

[0033] If the platform foreign object recognition result is that a platform foreign object is recognized in the area to be recognized, it further includes: Statistical information on the types of foreign objects and the location information of foreign objects.

[0034] It can be understood that by statistically analyzing the information on the types of foreign objects and the location information of foreign objects, the foreign object information is sorted and classified, which is convenient for the relearning of the foreign object recognition model and also convenient for the train crew to consult, further improving the safety of train operation.

[0035] To better implement the platform foreign object recognition method in the embodiments of the present invention, correspondingly, on the basis of the platform foreign object recognition method, as Figure 2 shown, the embodiments of the present invention further provide a platform foreign object recognition system. The platform foreign object recognition system 200 includes: A camera 201 for collecting the image information inside the platform in real time and transmitting the image information to a switch 202 through a network cable; A display control room 202 for transmitting the train entry signal to the switch 202 through an optical fiber; A switch 203 for transmitting the recognition signal and the image information to a server 204 after receiving the train entry signal; A server 204 for performing foreign object recognition on the image information through a preset foreign object recognition model after receiving the recognition signal; A display 205 for displaying the foreign object recognition result.

[0036] Among them, if the platform foreign object recognition result is that no platform foreign object is recognized in the area to be recognized, the server 204 recognizes the next frame of image information until the display control room 202 sends a signal indicating no need for detection; If the result of platform foreign object recognition is that a platform foreign object is recognized in the area to be recognized, when there is a foreign object, the server 204 sends out an alarm signal and a train suspension signal, and transmits the alarm signal and the train suspension signal to the switch 201 through the network cable, so that the switch 201 transmits the alarm signal and the train suspension signal to the display control room 202 through the optical fiber.

[0037] The platform foreign object recognition system 200 provided in the above embodiment can implement the technical solutions described in the above embodiment of the platform foreign object recognition method. The specific implementation principles of the above modules or units can be referred to the corresponding content in the above embodiment of the platform foreign object recognition method, which will not be elaborated here.

[0038] As Figure 3 shown, the present invention also correspondingly provides an electronic device 300. The electronic device 300 includes a processor 301, a memory 302 and a display 303. Figure 3 Only some components of the electronic device 300 are shown, but it should be understood that it is not required to implement all the shown components, and more or fewer components can be alternatively implemented.

[0039] In some embodiments, the processor 301 may be a central processing unit (CPU), a microprocessor or other data processing chips, and is used to run the program code stored in the memory 302 or process data, such as the platform foreign object recognition method in the present invention. (Note: The program code or process data of the present invention runs on the server. Since the data needs to be stored for 30 days and the data volume is relatively large, a separate memory is set, and the server can call the data in the memory later) In some embodiments, the processor 301 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, the processor 301 may be local or remote. In some embodiments, the processor 301 may be implemented on a cloud platform. In one embodiment, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-cloud, etc., or any combination of the above.

[0040] In some embodiments, the memory 302 may be an internal storage unit of the electronic device 300, such as the hard disk or memory of the electronic device 300. In some other embodiments, the memory 302 may also be an external storage device of the electronic device 300, such as a plug-in hard disk equipped on the electronic device 300, a smart media card (SMC), a secure digital (SD) card, a flash card, etc.

[0041] Further, the memory 302 may include both the internal storage unit of the electronic device 300 and external storage devices. The memory 302 is used to store the application software installed in the electronic device 300 and various types of data.

[0042] In some embodiments, the display 303 may be an LED display, a liquid crystal display, a touch liquid crystal display, and an OLED (Organic Light-Emitting Diode) toucher, etc. The display 303 is used to display the information of the electronic device 300 and to display a visual user interface. The components 301-303 of the electronic device 300 communicate with each other through the system bus.

[0043] In some embodiments, when the processor 301 executes the platform foreign object recognition program in the memory 302, the following steps may be implemented: Collect image information within the platform; Perform regional division on the image information to obtain the area to be recognized and the non-recognized area of the image information; Based on a pre-trained and complete foreign object recognition model, recognize the area to be recognized to obtain the platform foreign object recognition result.

[0044] It should be understood that when the processor 301 executes the platform foreign object recognition program in the memory 302, in addition to the above functions, other functions may also be implemented. For details, reference may be made to the description of the corresponding method embodiments above.

[0045] Further, the embodiments of the present invention do not specifically limit the type of the mentioned electronic device 300. The electronic device 300 may be a mobile phone, a tablet computer, a personal digital assistant (PDA), a wearable device, a laptop computer, etc. Exemplary embodiments of the portable electronic device include, but are not limited to, portable electronic devices equipped with IOS, android, microsoft, or other operating systems. The above portable electronic devices may also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the present invention, the electronic device 300 may not be a portable electronic device, but a desktop computer with a touch-sensitive surface (e.g., a touch panel).

[0046] Correspondingly, the embodiments of the present application also provide a computer-readable storage medium. The computer-readable storage medium is used to store computer-readable programs or instructions. When the programs or instructions are executed by a processor, the steps or functions in the platform foreign object recognition method provided by the above method embodiments can be implemented.

[0047] Those skilled in the art can understand that all or part of the processes of implementing the methods of the above embodiments can be completed by instructing relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. Among them, the computer-readable storage medium is a disk, an optical disc, a read-only memory or a random access memory, etc.

[0048] The above are only the preferred specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention.

Claims

1. A method for identifying foreign objects on a platform, characterized in that: include: Collect image information within the station; Divide the image information into regions to obtain a to-be-recognized region and a non-recognized region of the image information; The area to be identified is identified based on a preset foreign object identification model to obtain a platform foreign object identification result.

2. The platform foreign object identification method according to claim 1, characterized in that: Also includes: In response to the train departure signal, image information inside the platform is collected.

3. The platform foreign object identification method according to claim 1, characterized in that: The dividing the image information into regions to obtain the to-be-recognized region and the non-recognized region of the image information includes: Performing grayscale processing on the image information to obtain grayscale value data corresponding to the image; The image information is divided into regions based on a preset gray value range to obtain a to-be-recognized region and a non-recognized region of the image information.

4. The platform foreign object identification method according to claim 3, characterized in that: The generation process of the preset foreign body recognition model includes: Collecting historical image information under the platform background, and annotating the historical image information based on preset foreign objects to obtain an annotated image information set, wherein the preset foreign objects are small objects that are easily overlooked; The transfer learning model is used to learn the labeled image information set to generate a foreign object recognition model.

5. The platform foreign object identification method according to claim 1, characterized in that: Also includes: If the platform foreign object recognition result is that no platform foreign object is recognized in the to-be-recognized area, the next frame of camera image information in the platform is recognized until all cameras complete their patrol.

6. The platform foreign object identification method according to claim 1, characterized in that: Also includes: If the platform foreign object identification result is that the platform foreign object is identified in the to-be-identified area, an alarm signal is sent to the train, and after the danger is eliminated, a safety signal is sent to the train to allow the train to move.

7. The platform foreign object identification method according to claim 6, characterized in that: Also includes: If the platform foreign object recognition result is that a platform foreign object is recognized in the area to be recognized; Count the type and location information of foreign objects.

8. A platform safety area foreign body identification system, characterized in that: include: The camera is used to collect image information in the station in real time and transmit the image information to the switch via a network cable; Display and control room, used to transmit train entry signals to the switch via optical fiber; A switch, configured to transmit the identification signal and the image information to a server after receiving a train entry signal; The server is used to perform foreign object recognition on the image information through a preset foreign object recognition model after receiving the recognition signal; A display is used to display the foreign body recognition result.

9. An electronic device, characterized in that: comprising a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps of the platform foreign object identification method described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps of the platform foreign object identification method described in any one of claims 1 to 7.

Images