Fault detection method and system for stop lamp of vehicle at toe-in station

Abstract

The invention discloses a toe-in station vehicle brake lamp fault detection method and system. The system comprises an image collector, a photoelectric sensor, a toe-in station controller, a server and a client, wherein the image collector is used for collecting a vehicle image of a toe-in station; the photoelectric sensor is used for outputting an occupying signal; the toe-in station controller comprises a timing unit, and the timing unit is used for recording vehicle occupation time; when the server responds to the situation that the occupation time is larger than zero, an image collector is controlled to collect a vehicle image, and a brake lamp state recognition model of the server recognizes the state of a vehicle brake lamp based on the vehicle image obtained within the occupation time; the server stores vehicle brake lamp state information based on the recognized vehicle brake lamp state, and when the vehicle brake lamp state is a non-lighting state and the occupation time is ended, the vehicle brake state information is sent to a user corresponding to the client code; and the client is used for receiving and displaying the vehicle brake lamp state information.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle brake light fault detection based on a toe-in station on an automobile assembly line, and in particular to a vehicle brake light fault detection method and system for a toe-in station. Background Art

[0002] During the vehicle assembly process, due to improper installation of the brake light wiring or faults in the brake light source module itself, the brake light will not light up when the vehicle is braked after CP7 rolls off the assembly line. If this fault is not identified until the vehicle is tested at CP8, the first inspection pass rate will be reduced.

[0003] There are currently two ways to detect the brake light failure: the first is by manual visual inspection, in which a person visually checks between the hubs to see if the brake light is off when the brake is applied. However, manual visual inspection wastes man-hours and there is a risk of human omissions. The second is through vehicle electrical inspection. The vehicle electrical inspection method has two disadvantages. The first disadvantage is that the electrical inspection system will regard the detected brake light failure as an accidental accident and ignore it. The second disadvantage is that in the case where the brake light source module fails and the brake light does not light up, the electrical inspection system's detection data are all normal and cannot be identified. Summary of the invention

[0004] To solve at least one aspect of the above problems, the present invention provides a method for detecting a brake light fault in a toe-in station, comprising: outputting an occupancy signal through a photoelectric sensor when a vehicle is at least partially located in the toe-in station; using a timing unit to record the vehicle occupancy time, the timing unit starts timing in response to the initial moment of the occupancy signal, and the timing unit resets the occupancy time in response to the termination moment of the occupancy signal; capturing a current vehicle image in the process of the vehicle entering the toe-in station through an image collector when the occupancy time is greater than zero; using a brake light state recognition model to evaluate the vehicle brake light state based on the vehicle image within the occupancy time; when the vehicle brake light state is in a lit state, stopping collecting vehicle images, receiving a vehicle code through an assembly server, and storing vehicle brake light state information, the vehicle brake light state information including the vehicle code, the vehicle image and the vehicle brake light state; when the vehicle brake light state is in an unlit state and the occupancy time expires, obtaining a vehicle code and a client code corresponding to the vehicle code through the assembly server, storing the vehicle brake light state information, and sending the vehicle brake state information to a user corresponding to the client code; receiving and displaying the vehicle brake state information in real time through a front-end page.

[0005] Preferably, when the vehicle brake light is in an off state, a buzzer arranged at the toe-in station outputs a prompt audio.

[0006] Preferably, the method for constructing the brake light state recognition model includes: outputting an occupancy signal through a photoelectric sensor when the vehicle is at least partially located in the toe-in station; recording the occupancy time of the vehicle in the toe-in station in response to the received occupancy signal; when the occupancy time is greater than zero, controlling the image collector to output a vehicle image for the vehicle captured image of the toe-in station, and setting an image acquisition cycle so that the number of vehicle images captured within the occupancy time is greater than a preset image value; performing image acquisition on multiple vehicle models respectively to generate a vehicle image set; using image annotation software to annotate the collected vehicle image set to generate a Yolov5 data set; and using the Yolov5 model to train and generate a vehicle brake state recognition model based on the Yolov5 data set.

[0007] On the other hand, a vehicle brake light fault detection system for a toe-in station is provided, comprising: an image collector, the image collector is arranged on a gantry beam of the toe-in station, the image collector is used to collect vehicle images at the toe-in station; a photoelectric sensor, the photoelectric sensor outputs an occupancy signal in response to the vehicle being at least partially located at the toe-in station; a toe-in station controller, the toe-in station controller is electrically connected to the photoelectric sensor, the toe-in station controller receives the occupancy signal through the photoelectric sensor, the toe-in station controller comprises a timing unit, the timing unit is used to record the vehicle occupancy time, the timing unit starts timing in response to the initial moment of the occupancy signal, and the timing unit resets the occupancy time in response to the termination moment of the occupancy signal; a server, the server receives the occupancy time in real time through the toe-in station controller, the server responds to the occupancy time being greater than zero The server outputs an image acquisition instruction when the image acquisition instruction is used to control the image acquisition device to acquire a vehicle image. The server includes a brake light state recognition model, and the brake light state recognition model recognizes the vehicle brake light state based on the vehicle image acquired during the occupancy time. The server also includes a code receiving unit, and the code receiving unit is used to receive a vehicle code and a client code. The server stores the vehicle brake light state information based on the recognized vehicle brake light state, and the vehicle brake light state information includes the vehicle code, the vehicle brake light state and the vehicle image. When the vehicle brake light state is not lit and the occupancy time ends, the server determines the corresponding client code according to the vehicle code, and sends the vehicle brake state information to the user corresponding to the client code; the client is connected to the server in communication, and the client is used to receive and display the vehicle brake light state information.

[0008] Preferably, a buzzer is further included. When the vehicle brake light is not on and the occupancy time expires, the server outputs a fault instruction, and the toe station controller controls the buzzer to output a sound alarm in response to the fault instruction.

[0009] Preferably, the server further comprises a data storage module, and the data storage module is used to store the vehicle code and its corresponding vehicle image and recognition result.

[0010] Preferably, the server further comprises a front-end display unit, and the front-end display unit is used to display vehicle brake light status information of the current vehicle.

[0011] Preferably, the server terminates outputting the image acquisition instruction in response to receiving a vehicle brake light lighting signal.

[0012] The method and system for detecting the fault of the brake light of a vehicle in the toe-in station of the embodiment of the present invention have the following beneficial effects: the detection model does not require a GPU to run, the hardware price is low, and it can be directly run on a CPU computer, or even on a Raspberry Pi, avoiding the high price of a GPU server. It supports dynamic detection, that is, detection is performed during the movement of the vehicle. During the entire time period from when the vehicle enters the toe-in station to when the driver steps on the brake, the vehicle stops, and the multiple frames of images in this time period are continuously detected by cyclically executing the "capture-detection" operation until the brake light is detected to be on, otherwise it means that the brake light is not on after the brake is stepped on, which is unqualified; the recognition result is less affected by the change of the ambient light source. Based on the Yolov5 training recognition model, the training samples include images collected under various ambient light sources, and the recognition result is basically not affected by the ambient light source; compared with the traditional method of detecting the voltage / current value of the brake light part in the internal circuit of the vehicle, the visual detection result is more accurate, because the brake light may fail to light up due to the problem of the parts themselves even when the voltage / current value is normal. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to better understand the above and other purposes, features, advantages and functions of the present invention, reference may be made to the embodiments shown in the accompanying drawings. The same reference numerals in the accompanying drawings refer to the same components. It should be understood by those skilled in the art that the accompanying drawings are intended to schematically illustrate the preferred embodiments of the present invention and have no limiting effect on the scope of the present invention, and the components in the drawings are not drawn to scale.

[0014] Figure 1 A schematic flow chart of a method for detecting a brake light fault at a toe-in station according to an embodiment of the present invention is shown;

[0015] Figure 2 A schematic diagram showing the construction process of a brake light state recognition model for a toe-in station vehicle brake light fault detection method according to an embodiment of the present invention

[0016] Figure 3 A structural block diagram of a vehicle brake light fault detection system for a toe-in station according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0017] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0018] As used herein, the term "including" and its variations mean open inclusion, i.e., "including but not limited to". Unless otherwise stated, the term "or" means "and / or". The term "based on" means "based at least in part on". The terms "an example embodiment" and "an embodiment" mean "at least one example embodiment". The term "another embodiment" means "at least one additional embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0019] In order to at least partially solve one or more of the above problems and other potential problems, an embodiment of the present disclosure provides a method for detecting a brake light fault in a toe-in station vehicle, comprising:

[0020] Step S1, outputting an occupancy signal by a photoelectric sensor when the vehicle is at least partially located in a toe-in position.

[0021] Specifically, a photoelectric sensor is added to the equipment of the toe-in station. The photoelectric sensor is triggered as the front of the vehicle enters the toe-in station. The photoelectric sensor starts to output an occupancy signal until the vehicle completes the detection of the toe-in station and leaves.

[0022] Step S2, using a timing unit to record the vehicle occupancy time, the timing unit starts timing in response to the initial moment of the occupancy signal, and the timing unit resets the occupancy time in response to the termination moment of the occupancy signal.

[0023] Specifically, when the timing unit receives the occupancy signal output by the photoelectric sensor when it is triggered, it starts timing, and uses the data storage unit (Data Blocks, DB block) to save the occupancy time value at a DB point on the DB block. When the vehicle leaves the toe-in station, the photoelectric sensor stops outputting the occupancy signal, and the occupancy time is reset by clearing the time data point on the DB block. When the next vehicle enters the toe-in station, the timing unit can re-record the vehicle's occupancy time in the toe-in station.

[0024] For example, the toe-in station includes a PLC controller, the PLC controller is in communication with the photoelectric sensor, the PLC controller also includes a DB block for storing the occupancy time, the server is in communication with the PLC controller, and the Snap7 library is used to read the occupancy time of the vehicle head after entering the toe-in station from the characteristic DB point on the toe-in device PLC every second through the S7 protocol; if the occupancy time read in step S2 is not empty and is within the range of 0 to a certain time value T, then step S3 is executed; otherwise, the execution returns to step S2;

[0025] Step S3, when the occupation time is greater than zero, the current vehicle image in the process of the vehicle entering the toe-in station is captured by the image collector.

[0026] Specifically, the image collector uses Hikvision SDK, and the image collector is connected to the server for communication. When the occupancy time is greater than zero, it is determined that the occupancy time is not empty. The server calls Hikvision SDK to capture the current frame image of the vehicle entering the toe-in process to obtain the vehicle image.

[0027] Step S4: using a brake light state recognition model to evaluate the vehicle brake light state based on the vehicle image during the occupancy time.

[0028] Specifically, the server also includes a brake light status recognition model. The server calls the brake light status recognition model in ONNX format trained based on the Yolov5 model through the OpenCV library to identify whether the vehicle brake lights in the acquired vehicle image are on; if it is identified that the brake lights in the current frame image of the vehicle are on, then execute step S5, otherwise, execute step S41. Step S41 includes: using the Snap7 library, through the S7 protocol, reading the occupancy time value after the front of the vehicle begins to enter the toe-in station from the toe-in device PLC controller; if the read occupancy time is still within the range of 0 to a certain time value T (T depends on the value of the current vehicle's occupancy time), then return to execute step S3; otherwise, execute steps S5, S51 and S52 at the same time;

[0029] Step S5, when the vehicle brake light state is on, stop collecting vehicle images, receive the vehicle code through the assembly server, and store the vehicle brake light state information, the vehicle brake light state information includes the vehicle code, vehicle picture and vehicle brake light state.

[0030] Specifically, when the vehicle brake light is on, the server controls the image collector to stop collecting vehicle images, and obtains the current vehicle number of the toe-in station from an external server (such as an assembly server), saves it in the vehicle number field (i.e., vehicle code) of the current row in the recognition result table of the MySQL database, saves the recognition result (OK, i.e., the vehicle brake light is on) in the recognition result field of the current row, and saves the URL address of the recognition result image (i.e., vehicle image) in the image address field of the current row.

[0031] Step S51, when the vehicle brake light state is off and the occupancy time expires, the vehicle code and the client code corresponding to the vehicle code are obtained through the assembly server, the vehicle brake light state information is stored, and the vehicle brake state information is sent to the user corresponding to the client code.

[0032] Specifically, when the occupancy time expires, that is, the vehicle leaves the toe-in station, and the vehicle brake light status is still not lit, the server obtains the current vehicle number of the toe-in station from an external server (such as an assembly server), saves it in the vehicle number field (i.e., vehicle code) of the current row in the recognition result table of the MySQL database, saves the recognition result (NOK, i.e., the vehicle brake light status is not lit) in the recognition result field of the current row, and saves the URL address of the recognition result image (i.e., vehicle image) in the image address field of the current row. And further send the vehicle brake light status information to the user through the client code corresponding to the vehicle code. For example, if the client uses DingTalk, the server calls the DingTalk robot API to send the unqualified result to the relevant DingTalk group for reminder;

[0033] Step S6, receiving and displaying the vehicle braking status information in real time through the front-end page.

[0034] Specifically, the server updates the front-end web page information with the vehicle number (i.e., vehicle code) of the current vehicle, the recognition result (OK or NOK, i.e., the vehicle brake light status), the recognition result image (vehicle image), and other information.

[0035] In some embodiments, when the vehicle brake light is in an off state, a buzzer disposed at the toe station outputs a prompt audio.

[0036] Specifically, it also includes step S52, when the vehicle brake light is not on, the server sends a signal to the toe-mounted device PLC controller through the S7 protocol to trigger the buzzer on the device. The PLC controller controls the connected buzzer to play a preset audio in response to the received signal.

[0037] In some embodiments, a method for constructing a brake light state recognition model includes:

[0038] Step T1, outputting an occupancy signal by means of a photoelectric sensor when the vehicle is at least partially located at the toe-in position.

[0039] Specifically, a photoelectric sensor is added to the toe-in device. As soon as the front of the vehicle enters the toe-in station, the photoelectric sensor is triggered to start timing, and the duration value is saved at a certain DB point on the DB block until the vehicle leaves after completing the toe-in detection. If the photoelectric sensor has no signal, the duration value is empty.

[0040] Step T2, recording the occupancy time of the vehicle in the toe-in station in response to the received occupancy signal.

[0041] Specifically, using the Snap7 library and the S7 protocol, the occupancy time of the vehicle head after entering the toe-in station is read from the characteristic DB point on the toe-in equipment PLC every second; if the occupancy time read in step S2 is not empty and is within the range of 0 to a certain time value T, then step S3 is executed; otherwise, the process returns to step S2;

[0042] Step T3, when the occupancy time is greater than zero, the image collector is controlled to output the vehicle image of the toe-in station, and the image acquisition cycle is set so that the number of vehicle images collected within the occupancy time is greater than the preset image value.

[0043] Specifically, the camera SDK is called to collect one vehicle image per second when the vehicle enters the toe-in process. In some embodiments, the image collection frequency is set so that the vehicle collects enough vehicle images during the occupation time of the toe-in station. In other embodiments, the occupation time of the vehicle in the toe-in station is controlled so that the vehicle is removed from the toe-in station when the number of vehicle images collected is greater than or equal to a preset image value (for example, 300 images).

[0044] Step T4, collecting images of multiple vehicle models respectively to generate a vehicle image set.

[0045] Specifically, vehicle images are collected for each of the multiple vehicle models, specifically by cyclically executing steps S2 and S3, so that more than 300 images of each vehicle model in the toe-in process are collected.

[0046] Step T5: Use image annotation software to annotate the collected vehicle image set to generate the Yolov5 data set.

[0047] Specifically, the image annotation software uses Labelimg software, and the collected images are manually annotated using Labelimg software to create the Yolov5 data set.

[0048] Step T6, using the Yolov5 model to train and generate a vehicle braking state recognition model based on the Yolov5 data set.

[0049] Specifically, using the Pytorch framework, a pt format recognition model covering whether the brake lights of all current car models are on or off is trained based on the Yolov5 model. That is, the Yolov5 model outputs a brake status recognition model in pt format.

[0050] In some embodiments, step T7 is also included, converting the trained pt format recognition model into ONNX format to facilitate direct calling of the OpenCV library;

[0051] On the other hand, a system for detecting brake light faults of vehicles at a toe-in station is provided, comprising: an image collector, a photoelectric sensor, a toe-in station controller, a server and a client, wherein the image collector is arranged on a gantry beam of the toe-in station, and is used to collect vehicle images at the toe-in station; the photoelectric sensor outputs an occupancy signal in response to the vehicle being at least partially located at the toe-in station; the toe-in station controller is electrically connected to the photoelectric sensor, and the toe-in station controller receives the occupancy signal through the photoelectric sensor; the toe-in station controller comprises a timing unit, and the timing unit is used to record the vehicle occupancy time, and the timing unit starts timing in response to the initial moment of the occupancy signal, and resets the occupancy time in response to the termination moment of the occupancy signal; the server receives the occupancy time in real time through the toe-in station controller, and the server responds to the occupancy time being greater than The image acquisition instruction is output at zero time, and the image acquisition instruction is used to control the image collector to acquire vehicle images. The server includes a brake light state recognition model, and the brake light state recognition model recognizes the vehicle brake light state based on the vehicle image acquired during the occupancy time. The server also includes a code receiving unit, and the code receiving unit is used to receive the vehicle code and the client code. The server stores the vehicle brake light state information based on the recognized vehicle brake light state. The vehicle brake light state information includes the vehicle code, the vehicle brake light state and the vehicle image. When the vehicle brake light state is not lit and the occupancy time ends, the server determines the corresponding client code according to the vehicle code, and sends the vehicle brake state information to the user corresponding to the client code; the client is connected to the server for communication, and the client is used to receive and display the vehicle brake light state information.

[0052] Specifically, the image collector uses a Hikvision network camera DS-2CD2T56WD-I34mm, and the image collector is installed on the gantry crossbeam of the toe-in station through a bracket to ensure that it can take pictures of the rear of the vehicle when the vehicle enters the toe-in station; the camera is connected to the server through a network cable.

[0053] The photoelectric sensor is connected to the toe station controller to provide an occupancy signal from the time the front part of the vehicle enters the toe station to the time the toe detection is completed and the vehicle leaves the toe station. For example, the brand of the photoelectric sensor is SICK, and the model is 1041376 WTB11-2P2431.

[0054] The toe-in station controller adopts a PLC controller, and the photoelectric sensor realizes communication connection with the toe-in station controller through the IO module of the toe-in equipment PLC.

[0055] The server can be a PC with a window system, a PC with a Linux system, or a Raspberry Pi; the server does not need to have an independent graphics card; a database is installed on the server. In some embodiments, the server is an ordinary computer with a Windows 10 system without an independent graphics card. The database installed on the server is a MySQL database. Snap7 and OpenCV libraries are installed on the server. The Snap7 library is used to communicate with the toe station controller PLC to read the current vehicle occupancy time; the OpenCV library is used to call the brake light state recognition model, which is an ONNX visual recognition model.

[0056] The server communicates with the toe-in station controller PLC to read the occupancy time value of the current vehicle in the PLC; drives the camera to capture images; runs the brake light status recognition model; determines whether the brake light is faulty based on the read current vehicle occupancy time value and the recognition result of whether the brake light is on or off, and determines that the vehicle's brake light is faulty when the brake light is not on; the server communicates with an external server (for example, an assembly server) to obtain the vehicle code of the current vehicle; stores the vehicle code of the current vehicle and the recognition result (qualified or unqualified) and the URL address of the recognized vehicle image as the same row in the database; acts as a database server to store the recognition results of each vehicle (vehicle code, vehicle image, brake light status); if a brake light failure is identified, calls the DingTalk robot to push the faulty vehicle number to the DingTalk group; if a brake light failure is identified, sends a signal to the toe-in device PLC to trigger the buzzer on the device

[0057] The client uses DingTalk client, and users register and log in on the client through the client code. Receiving the vehicle brake light status information of the faulty vehicle through the client can timely and accurately remind the corresponding technicians of the vehicle, so as to further deal with the faulty vehicle in time.

[0058] In some embodiments, a buzzer is further included. When the vehicle brake light is not on and the occupancy time expires, the server outputs a fault instruction, and the toe station controller controls the buzzer to output a sound alarm in response to the fault instruction.

[0059] Specifically, the buzzer is connected to the toe-in station controller for communication. When the vehicle brake light fails, the toe-in station controller can be used to control the buzzer for prompting. In another embodiment, the toe-in station controller includes a first audio and a second audio. When the vehicle brake light is not lit and the occupancy time expires, the server outputs a fault instruction and the buzzer plays the first audio. When the vehicle brake light is lit, the server outputs an instruction and the buzzer plays the second audio.

[0060] In some embodiments, the server further includes a data storage module, which is used to store the vehicle code and its corresponding vehicle image and recognition result.

[0061] Specifically, the server stores the vehicle brake light status information through the data storage module for easy management and query.

[0062] In some embodiments, the server also includes a front-end display unit, which is used to display vehicle brake light status information of the current vehicle.

[0063] Specifically, on the server, the VUE front-end framework and the egg.js back-end framework are used to develop a web page that displays the recognition results of the vehicle brake light status information. The server acts as a web server to generate a web page that updates the current vehicle brake light recognition results, vehicle number, and the recognized image in real time for external browsers to query and access;

[0064] In some embodiments, the server terminates outputting the image acquisition instruction in response to receiving a vehicle brake light on signal.

[0065] Specifically, when the brake lights of the vehicle are on, it is determined that the brake lights of the vehicle are in normal state, and further driving of the image collector is avoided by terminating the output of the image acquisition instruction, thereby reducing the occupancy of storage resources and reducing the computing cost.

[0066] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements in the market, or to enable other persons of ordinary skill in the art to understand the present invention.

Claims

1. A method for detecting vehicle brake light faults at a toe-in station, characterized in that: include: Outputting an occupancy signal by means of a photoelectric sensor when the vehicle is at least partially located in the toe-in position; A timing unit is used to record the vehicle occupancy time, wherein the timing unit starts timing in response to the initial moment of the occupancy signal, and the timing unit resets the occupancy time in response to the termination moment of the occupancy signal; When the occupation time is greater than zero, the current vehicle image in the process of the vehicle entering the toe-in station is captured by the image collector; The brake light state recognition model is used to evaluate the vehicle brake light state based on the vehicle image during the occupancy time; When the vehicle brake light is in the on state, stop collecting vehicle images, receive the vehicle code through the assembly server, and store the vehicle brake light state information, wherein the vehicle brake light state information includes the vehicle code, the vehicle image, and the vehicle brake light state; When the vehicle brake light state is off and the occupancy time expires, the vehicle code and the client code corresponding to the vehicle code are obtained through the assembly server, the vehicle brake light state information is stored, and the vehicle brake state information is sent to the user corresponding to the client code; The vehicle braking status information is received and displayed in real time through the front-end page.

2. The method according to claim 1, characterized in that When the vehicle brake light is not on, a buzzer arranged at the toe-in station outputs a prompt audio.

3. The method according to claim 2, characterized in that The method for constructing the brake light state recognition model includes: Outputting an occupancy signal by means of a photoelectric sensor when the vehicle is at least partially located in the toe-in position; recording the occupancy time of the vehicle at the toe-in station in response to the received occupancy signal; When the occupancy time is greater than zero, the image collector is controlled to output the vehicle image of the toe-in station, and the image acquisition cycle is set so that the number of vehicle images collected within the occupancy time is greater than the preset image value; Collect images of multiple vehicle models respectively to generate a vehicle image set; Use image annotation software to annotate the collected vehicle image set and generate the Yolov5 dataset; The Yolov5 model is used to train and generate a vehicle braking state recognition model based on the Yolov5 dataset.

4. A vehicle brake light fault detection system for a toe-in station, characterized in that: include: An image collector, the image collector is arranged on the gantry crossbeam of the toe-in station, and the image collector is used to collect vehicle images at the toe-in station; a photoelectric sensor, the photoelectric sensor outputting an occupancy signal in response to the vehicle being at least partially located in the toe-in position; a toe-in station controller, the toe-in station controller being electrically connected to the photoelectric sensor, the toe-in station controller receiving the occupancy signal through the photoelectric sensor, the toe-in station controller comprising a timing unit, the timing unit being used to record the vehicle occupancy time, the timing unit starting timing in response to the initial moment of the occupancy signal, and the timing unit resetting the occupancy time in response to the termination moment of the occupancy signal; a server, wherein the server receives the occupancy time in real time through the toe-in station controller, and the server outputs an image acquisition instruction in response to the occupancy time being greater than zero, wherein the image acquisition instruction is used to control the image collector to acquire a vehicle image, wherein the server includes a brake light state recognition model, wherein the brake light state recognition model recognizes the vehicle brake light state based on the vehicle image acquired during the occupancy time, wherein the server also includes a code receiving unit, wherein the code receiving unit is used to receive a vehicle code and a client code, wherein the server stores vehicle brake light state information based on the recognized vehicle brake light state, wherein the vehicle brake light state information includes a vehicle code, a vehicle brake light state, and a vehicle image, and wherein when the vehicle brake light state is not lit and the occupancy time expires, the server determines a corresponding client code according to the vehicle code, and sends the vehicle brake state information to a user corresponding to the client code; A client is connected to the server for communication, and is used for receiving and displaying the vehicle brake light status information.

5. The system according to claim 4, characterized in that It also includes a buzzer. When the vehicle brake light is in an off state and the occupancy time expires, the server outputs a fault instruction, and the toe station controller controls the buzzer to output a sound alarm in response to the fault instruction.

6. The system according to claim 5, characterized in that The server also includes a data storage module, which is used to store vehicle codes and their corresponding vehicle images and recognition results.

7. The system according to claim 6, characterized in that The server also includes a front-end display unit, which is used to display vehicle brake light status information of the current vehicle.

8. The system according to claim 7, characterized in that The server terminates outputting the image acquisition instruction in response to receiving the vehicle brake light lighting signal.

Images