Vehicle movement detection method and terminal in car washing process

Abstract

The present invention discloses a vehicle movement detection method and terminal in a car wash process, comprising the following steps: obtaining first video information shot by a first camera of sample vehicles of different vehicle sizes located in a car wash area during the car wash process and second video information shot when the vehicle is not washed, obtaining first picture information, and simulating third picture information of each openable component of each sample vehicle in an open state during the car wash process, thereby training and generating an image classification model to monitor whether the openable components of the vehicle are opened by mistake; and realizing live intrusion detection and vehicle movement detection through a first alarm detector and a second alarm detector, thereby obtaining the final monitoring result. The present invention can realize monitoring and identification of intrusion and vehicle misoperation in a car wash process, and its implementation cost is low and its accuracy is high.

Description

[0001] This case is a divisional application based on the invention patent with application date of January 12, 2021, application number 202110035208.7, and name “Video surveillance method and terminal for car washing process” as the parent case. Technical Field

[0002] The present invention relates to the technical field of vehicle monitoring, and in particular to a vehicle movement detection method and a terminal during a car washing process. Background Art

[0003] As the number of cars increases year by year, the demand for car-related services is also growing. Among them, car washing service is one of the essential car services for people's daily use of cars. Whether it is wind, sand, mud or dust, car owners have to wash their cars every once in a while.

[0004] Existing car wash services include manual washing and automatic washing. During the car washing process in an automatic car wash system, on the one hand, there is a situation where a person accidentally breaks into the automatic car wash system. For example, if the automatic car wash system does not perform any action, but may be waiting or entering the next program, etc., if a person enters the automatic car wash system at this time, it is easy to cause harm to the person; on the other hand, a person may perform an erroneous operation in the vehicle, such as accidentally pressing the button to open the door, etc. If the automatic car wash system cannot respond in time, it will cause damage to the person or the vehicle. Therefore, the existing automatic car wash machine lacks a method that can effectively monitor and identify intrusions and vehicle erroneous operations during the car wash process. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a vehicle movement detection method and a terminal during a car washing process, which can monitor and identify the vehicle movement during the car washing process.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0007] The vehicle movement detection method in the car washing process comprises the following steps:

[0008] S1, obtaining first video information shot by a first camera of sample vehicles of different vehicle sizes located in the car wash area during a car wash process and second video information shot when the vehicles are not washed, wherein the first camera is used to monitor all openable parts of the sample vehicles located in the car wash area, and all openable parts of the sample vehicles in the first video information are in a closed state, and the second video information includes video information of each openable part of the sample vehicles being opened in turn, and the openable parts include windows, sunroofs, doors and trunks;

[0009] S2, obtaining first picture information of each sample vehicle at each car washing step from the first video information, and marking each first picture information as a normal vehicle and adding it to the training image set;

[0010] S3, obtaining second picture information of each openable component of each sample vehicle in an open state from the second video information, simulating third picture information of each openable component of each sample vehicle in an open state during a car wash process based on the first picture information and the second picture information, and marking each piece of the third picture information as a vehicle abnormality and adding it to the training image set;

[0011] S4, training an image classification model capable of distinguishing between a normal vehicle and an abnormal vehicle according to the training image set;

[0012] S5, obtaining a monitoring video of a first vehicle in a car washing process shot in real time by the first camera, and obtaining a frame of monitoring image to be judged from the monitoring video at a first preset time interval as an input of the trained image classification model to obtain a first monitoring result of the image classification model, wherein the first monitoring result is that the vehicle is normal or abnormal;

[0013] S6, obtaining a second monitoring result of the first alarm detector performing live intrusion detection on the entrance and exit of the car wash area, and obtaining a third monitoring result of the second alarm detector performing movement detection on the first vehicle in the car wash area, wherein the second monitoring result is that no live intrusion is found or that live intrusion is found, and the third monitoring result is that the vehicle is normal or abnormal;

[0014] S7. Summarize the first monitoring result, the second monitoring result and the third monitoring result. If any one of the first monitoring result and the third monitoring result indicates that the vehicle is abnormal or the second monitoring result indicates that a living body is found to be intruding, an abnormal car wash alarm prompt message is issued.

[0015] In order to solve the above technical problems, another technical solution adopted by the present invention is:

[0016] The vehicle movement detection terminal for a car washing process comprises 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 vehicle movement detection method for a car washing process is implemented.

[0017] The beneficial effects of the present invention are as follows: the vehicle movement detection method and terminal in the car wash process, by shooting videos of the car wash process when the openable parts of sample vehicles of different vehicle sizes located in the car wash area are in a closed state and the unwashed process when the openable parts are in an open state, and then obtaining the corresponding image information from the two videos respectively, and then simulating the third image information of each openable part of each sample vehicle in an open state during the car wash process based on the two image information, that is, it is not necessary to open the openable parts of the vehicle during the car wash process to obtain the image of the openable parts of the vehicle in the open state in the car wash scene, thereby obtaining a lower cost and higher accuracy image classification model to identify vehicle misoperation. Afterwards, the first alarm detector and the second alarm detector are used to monitor intrusion and vehicle movement, so as to realize the monitoring and identification of intrusion and vehicle misoperation in the car wash process, and the implementation cost is low and the accuracy is high. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flow chart of a vehicle movement detection method in a car washing process according to an embodiment of the present invention;

[0019] Figure 2 is a schematic structural diagram of a second alarm detector in an initial state involved in an embodiment of the present invention;

[0020] Figure 3 It is a structural schematic diagram of the second alarm detector in the monitoring state involved in the embodiment of the present invention;

[0021] Figure 4 The figure is a schematic diagram of the structure of a vehicle movement detection terminal in a car washing process according to an embodiment of the present invention.

[0022] Description of labels:

[0023] 1. Vehicle movement detection terminal during car washing process; 2. Processor; 3. Memory;

[0024] 4. First induction member; 5. First rotating shaft; 6. Second induction member; 7. Second rotating shaft; 8. First vehicle. DETAILED DESCRIPTION

[0025] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in conjunction with the implementation modes and the accompanying drawings.

[0026] Please refer to Figures 1 to 3 , a vehicle movement detection method in a car washing process comprises the following steps:

[0027] S1, obtaining first video information shot by a first camera of sample vehicles of different vehicle sizes located in the car wash area during a car wash process and second video information shot when the vehicles are not washed, wherein the first camera is used to monitor all openable parts of the sample vehicles located in the car wash area, and all openable parts of the sample vehicles in the first video information are in a closed state, and the second video information includes video information of each openable part of the sample vehicles being opened in turn, and the openable parts include windows, sunroofs, doors and trunks;

[0028] S2, obtaining first picture information of each sample vehicle at each car washing step from the first video information, and marking each first picture information as a normal vehicle and adding it to the training image set;

[0029] S3, obtaining second picture information of each openable component of each sample vehicle in an open state from the second video information, simulating third picture information of each openable component of each sample vehicle in an open state during a car wash process based on the first picture information and the second picture information, and marking each piece of the third picture information as a vehicle abnormality and adding it to the training image set;

[0030] S4, training an image classification model capable of distinguishing between a normal vehicle and an abnormal vehicle according to the training image set;

[0031] S5, obtaining a monitoring video of a first vehicle in a car washing process shot in real time by the first camera, and obtaining a frame of monitoring image to be judged from the monitoring video at a first preset time interval as an input of the trained image classification model to obtain a first monitoring result of the image classification model, wherein the first monitoring result is that the vehicle is normal or abnormal;

[0032] S6, obtaining a second monitoring result of the first alarm detector performing live intrusion detection on the entrance and exit of the car wash area, and obtaining a third monitoring result of the second alarm detector performing movement detection on the first vehicle in the car wash area, wherein the second monitoring result is that no live intrusion is found or that live intrusion is found, and the third monitoring result is that the vehicle is normal or abnormal;

[0033] S7. Summarize the first monitoring result, the second monitoring result and the third monitoring result. If any one of the first monitoring result and the third monitoring result indicates that the vehicle is abnormal or the second monitoring result indicates that a living body is found to be intruding, an abnormal car wash alarm prompt message is issued.

[0034] From the above description, it can be seen that the beneficial effect of the present invention is that: by taking videos of the car washing process when the openable parts of sample vehicles of different vehicle sizes located in the car washing area are in a closed state and the non-car washing process when the openable parts are in an open state, and then obtaining the corresponding image information from the two videos respectively, and then simulating the third image information of each openable part of each sample vehicle in an open state during the car washing process based on the two image information, that is, there is no need to open the openable parts of the vehicle during the car washing process to obtain the image of the openable parts of the vehicle in the open state in the car washing scene, thereby obtaining a lower cost and higher accuracy image classification model to identify vehicle misoperation. Afterwards, the first alarm detector and the second alarm detector are used to monitor intrusion and vehicle movement, so as to realize the monitoring and identification of intrusion and vehicle misoperation during the car washing process, which has a low cost and high accuracy.

[0035] Furthermore, in step S1, obtaining the second video information shot by the first camera of sample vehicles of different vehicle sizes located in the car wash area before washing the vehicle specifically includes the following steps:

[0036] Classify all sample vehicles according to the categories and quantities of the openable parts to obtain all vehicle categories and the corresponding openable parts;

[0037] Arrange and combine the states of all the openable components in each of the vehicle categories to obtain a first set for each of the vehicle categories, wherein the state of each of the openable components includes a closed state and an open state, and at least one of the elements in the first set is in an open state;

[0038] Compile a car wash control program for each vehicle category according to the first set of each vehicle category, and the car wash control program responds to each element in the first set in sequence according to the order of the first set;

[0039] Establishing a communication connection with the sample vehicle, using a corresponding vehicle control program according to the vehicle category corresponding to the sample vehicle located in the car wash area, and loading the corresponding vehicle control program into the sample vehicle, so that the sample vehicle performs vehicle control according to the vehicle control program;

[0040] The second video information captured by the first camera when executing the corresponding automobile control program on sample vehicles of different vehicle sizes located in the car wash area is obtained.

[0041] From the above description, it can be seen that the images to be collected are pre-written into a car wash control program and loaded into the sample vehicle, and the vehicle is controlled according to the car wash control program, so that different sample vehicles control the openable parts in the same order. On the one hand, the program automatic control is more accurate, and the program can automatically mark the second image information without manual marking, and the image data required for training can be obtained. On the other hand, including the arrangement and combination of all openable parts, it is ensured that a sufficient number of image data with differences can be obtained for training, so that the image classification model can be trained quickly and accurately.

[0042] Furthermore, the step S3 specifically includes the following steps:

[0043] Second picture information of each openable component of each sample vehicle in an open state is obtained from the second video information, and the sample vehicles with the openable components in the open state in the second picture information and the car wash scenes other than the sample vehicles in the first picture information are fused to obtain third picture information of each openable component of each sample vehicle in an open state during the car wash process, and each piece of the third picture information is marked as a vehicle abnormality and then added to the training image set.

[0044] From the above description, it can be seen that the first image information and the second image information can be fused by using Photoshop or other P-image software to simulate and generate the third image information that conforms to the real scene to train the image classification model.

[0045] Furthermore, the step S3 specifically includes the following steps:

[0046] Acquire second picture information of each of the openable components of each of the sample vehicles in an open state from the second video information;

[0047] Pre-establishing a car wash simulation scene that restores the car wash area and the car wash process and a vehicle 3D model of each of the sample vehicles, wherein the simulated components corresponding to the openable components in the vehicle 3D model can be opened according to the openable components;

[0048] Restore a real-time car wash scene according to the first picture information and the car wash simulation scene, restore a real-time vehicle model according to the second picture information and the vehicle 3D model, and after placing the real-time vehicle model into the real-time car wash scene, obtain a real-time fusion scene for each of the sample vehicles;

[0049] The real-time fusion scene is photographed according to the virtual camera corresponding to the first camera in the real-time fusion scene to obtain third picture information of each of the openable parts of each sample vehicle in an open state during the car washing process, and each piece of the third picture information is marked as a vehicle abnormality and then added to the training image set.

[0050] From the above description, it can be seen that by establishing a vehicle 3D model and a real-time car wash scene to restore the real-time fusion scene, the third picture information that conforms to the real scene is captured to train the image classification model.

[0051] Furthermore, the first alarm detector is a planar alarm detector, and the second alarm detector is a linear alarm detector.

[0052] Furthermore, both the first alarm detector and the second alarm detector are dual-technology detectors with passive infrared detection and microwave detection.

[0053] Furthermore, the first alarm detector is a dual-technology detector with passive infrared detection and microwave detection, and the second alarm detector is a forward tilt sensor.

[0054] Furthermore, the first alarm detector and the second alarm detector are both first cameras, and step S6 specifically includes the following steps:

[0055] Obtaining a real-time monitoring image of the entrance and exit of the car wash area and the car wash area by the first camera, and determining whether two real-time monitoring images separated by a second preset time have differences at the entrance and exit of the car wash area; if so, generating a second monitoring result that a living body intrusion is found; otherwise, generating a second monitoring result that no living body intrusion is found;

[0056] It is determined whether there is a difference between the two real-time monitoring images separated by a second preset time on the car wash area. If so, a third monitoring result of a normal vehicle is generated; otherwise, a third monitoring result of a normal vehicle is generated.

[0057] From the above description, it can be seen that different usage requirements can be met by different selections of the first alarm detector and the second alarm detector. For example, all monitoring is performed through the first camera, which makes the equipment implementation simpler, but has higher requirements for software processing; while the use of linear alarm detectors is sensitive and the software processing is relatively simple, so the corresponding first alarm detector and second alarm detector can be selected according to the actual usage scenario.

[0058] Further, if there is one first camera, the first camera is located in the upper middle of the rear of the vehicle to be washed in the car washing area;

[0059] If there are two first cameras, the two first cameras are located above both ends of the diagonal line of the vehicle to be cleaned;

[0060] If there are three first cameras, the three first cameras are respectively located above both sides and the rear side of the vehicle to be cleaned.

[0061] It can be seen from the above description that, with different numbers of cameras, further limitations are made on the arrangement positions of different cameras to ensure that all openable parts of the vehicle to be washed in the car wash area can be monitored in real time.

[0062] Please refer to Figure 4 A vehicle movement detection terminal for a car washing process includes 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 vehicle movement detection method for a car washing process is implemented.

[0063] From the above description, it can be seen that the beneficial effect of the present invention is that: by taking videos of the car washing process when the openable parts of sample vehicles of different vehicle sizes located in the car washing area are in a closed state and the non-car washing process when the openable parts are in an open state, and then obtaining the corresponding image information from the two videos respectively, and then simulating the third image information of each openable part of each sample vehicle in an open state during the car washing process based on the two image information, that is, there is no need to open the openable parts of the vehicle during the car washing process to obtain the image of the openable parts of the vehicle in the open state in the car washing scene, thereby obtaining a lower cost and higher accuracy image classification model to identify vehicle misoperation. Afterwards, the first alarm detector and the second alarm detector are used to monitor intrusion and vehicle movement, so as to realize the monitoring and identification of intrusion and vehicle misoperation during the car washing process, which has a low cost and high accuracy.

[0064] Please refer to Figure 1 , Embodiment 1 of the present invention is:

[0065] The vehicle movement detection method in the car washing process comprises the following steps:

[0066] S1, obtaining first video information shot by a first camera of sample vehicles of different vehicle sizes located in a car wash area during a car wash process and second video information shot when the vehicles are not washed, wherein the first camera is used to monitor all openable parts of the sample vehicles located in the car wash area, wherein all openable parts of the sample vehicles in the first video information are in a closed state, and the second video information includes video information of each openable part of the sample vehicles being in an open state in turn, wherein the openable parts include windows, sunroofs, doors, and trunks;

[0067] In this embodiment, different vehicle types refer to the size of the vehicle, and can be classified according to a preset length and width range.

[0068] In this embodiment, there are three first cameras, which are respectively located above the two sides and the rear side of the vehicle to be washed, so that the openable parts on the entire vehicle to be washed can be fully monitored. In other equivalent embodiments, if there is one first camera, the first camera is located in the middle and upper part of the rear of the vehicle to be washed in the car washing area; if there are two first cameras, the two first cameras are located above the two ends of the diagonal of the vehicle to be washed, that is, it is ensured that the openable parts on the entire vehicle to be washed can be fully monitored.

[0069] In this embodiment, all the openable parts in the first video information of the vehicle in the car wash process are in the closed state, so it is only necessary to drive the sample vehicle into the car wash area.

[0070] As for the openable parts being in the open state, there are many openable parts, some are open and some are not open, and there are many such combinations. If it is controlled manually, it is definitely impossible to fully cover all situations. Therefore, in this embodiment, the second video information shot by the first camera of the sample vehicles of different vehicle sizes located in the car wash area before the car wash in step S1 specifically includes the following steps:

[0071] S11, classifying all sample vehicles according to the types and quantities of openable parts, and obtaining all vehicle types and corresponding openable parts;

[0072] Among them, in actual application scenarios, normal vehicles all have a sunroof, a trunk, four doors in the front, rear, left and right, and corresponding four windows. Some vehicles do not have a sunroof or have only two doors and windows. Therefore, they are classified according to the type and quantity of openable parts, that is, the type and quantity of openable parts in the same category are in a consistent corresponding relationship.

[0073] S12, arranging and combining the states of all openable components in each vehicle category to obtain a first set for each vehicle category, wherein the state of each openable component includes a closed state and an open state, and at least one openable component in any element of the first set is in an open state;

[0074] In this embodiment, for example, this category of vehicles has a sunroof, a trunk, four doors on the front, rear, left and right sides, and corresponding four windows, so there are a total of 10 openable components, of which the number of elements with only one openable component is 10, the number of elements with only two openable components is 45, and so on, the number of elements in the first set is thousands, and all the combined elements are written in a preset order through the program.

[0075] S13, compiling a car wash control program for each vehicle category according to the first set for each vehicle category, wherein the car wash control program responds to each element in the first set in sequence according to the order of the first set;

[0076] S14, establishing a communication connection with the sample vehicle, using a corresponding vehicle control program according to the vehicle category corresponding to the sample vehicle located in the car wash area, and loading the corresponding vehicle control program into the sample vehicle, so that the sample vehicle performs vehicle control according to the vehicle control program;

[0077] S15. Obtain second video information captured by the first camera when executing a corresponding vehicle control program on sample vehicles of different vehicle sizes located in the car wash area.

[0078] Since the control order of different sample vehicles in the same category is the same, we can know which element of the car wash control program the current sample vehicle is in based on the time point in the video, so we can perform automatic labeling and obtain a large number of effective training images.

[0079] S2, obtaining the first picture information of each sample vehicle in each car washing step from the first video information, and marking each first picture information as a normal vehicle and adding it to the training image set;

[0080] S3, obtaining second picture information of each openable component of each sample vehicle in an open state from the second video information, simulating third picture information of each openable component of each sample vehicle in an open state during the car wash process based on the first picture information and the second picture information, and marking each piece of third picture information as a vehicle abnormality and adding it to the training image set;

[0081] In this embodiment, step S3 specifically includes the following steps:

[0082] Second picture information of each openable component of each sample vehicle in an open state is obtained from the second video information, and image fusion is performed on the sample vehicles with openable components in an open state in the second picture information and the car wash scene other than the sample vehicles in the first picture information to obtain third picture information of each openable component of each sample vehicle in an open state during the car wash process, and each piece of third picture information is marked as a vehicle abnormality and then added to the training image set.

[0083] In other equivalent embodiments, step S3 specifically includes the following steps:

[0084] Acquire second picture information of each openable component of each sample vehicle in an open state from the second video information;

[0085] A car wash simulation scene restoring a car wash area and a car wash process and a 3D vehicle model of each sample vehicle are pre-established, and the simulated parts corresponding to the openable parts in the vehicle 3D model can be opened according to the openable parts;

[0086] A real-time car wash scene is restored based on the first image information and the car wash simulation scene, a real-time vehicle model is restored based on the second image information and the vehicle 3D model, and after the real-time vehicle model is placed in the real-time car wash scene, a real-time fusion scene of each sample vehicle is obtained;

[0087] The real-time fusion scene is photographed according to the virtual camera corresponding to the first camera in the real-time fusion scene to obtain third picture information of each openable component of each sample vehicle in an open state during the car washing process, and each piece of third picture information is marked as a vehicle abnormality and then added to the training image set.

[0088] Of course, no matter which of the above image fusion is based on, since the number of second image information itself is large, and there is at least one first image information in different car washing steps, the number of third image information finally formed will reach more than 10,000 levels, which can ensure that the trained image classification model has high accuracy and reliability.

[0089] S4, training an image classification model capable of distinguishing between normal vehicles and abnormal vehicles according to the training image set;

[0090] In this embodiment, the image classification model can not only distinguish between normal and abnormal vehicles, but also arrange and combine the states of all openable parts based on the above, so that when marking, it can be marked which openable parts are in the open state and which are in the closed state. In this way, a separate classification model for each openable part can be trained, and then data can be summarized based on the separate classification model of each openable part. Finally, not only can the abnormal state of the vehicle be obtained, but also which openable part is opened can be known.

[0091] Among them, all the third picture information about the currently trained openable component that is marked as open is marked as abnormal, and all the other third picture information is marked as normal, and then put into an original convolutional neural network model for training to obtain a separate classification model for this openable component. For example, if it is a separate classification model for the sunroof, then in the separate classification model for the sunroof, whether the trunk, door or window is in an open state or a closed state does not affect the final result. Only when the sunroof is in an open state, the separate classification model of the sunroof outputs an abnormality, otherwise it is normal.

[0092] S5, obtaining a monitoring video of a first vehicle in a car washing process shot in real time by a first camera, and obtaining a frame of a monitoring image to be judged from the monitoring video at a first preset time interval as an input of the trained image classification model to obtain a first monitoring result of the image classification model, wherein the first monitoring result is that the vehicle is normal or abnormal;

[0093] In this embodiment, the first preset time may be the time of one frame, that is, one sixteenth of a second, which may be comprehensively considered based on the required accuracy and processing speed, but the interval of the first preset time is (0, 1 second].

[0094] S6, obtaining a second monitoring result of the first alarm detector performing live intrusion detection on the entrance and exit of the car wash area, and obtaining a third monitoring result of the second alarm detector performing movement detection on the first vehicle in the car wash area, wherein the second monitoring result is that no live intrusion is found or that live intrusion is found, and the third monitoring result is that the vehicle is normal or that the vehicle is abnormal;

[0095] In this embodiment, the first alarm detector and the second alarm detector are both first cameras, and the above step S6 specifically includes the following steps:

[0096] Obtaining real-time monitoring images of the entrance and exit of the car wash area and the car wash area by the first camera, and determining whether two real-time monitoring images separated by a second preset time have differences at the entrance and exit of the car wash area; if so, generating a second monitoring result that finds a living body intrusion; otherwise, generating a second monitoring result that does not find a living body intrusion;

[0097] It is determined whether there is a difference between the two real-time monitoring images separated by a second preset time in the car wash area. If so, a third monitoring result of a normal vehicle is generated; otherwise, a third monitoring result of a normal vehicle is generated.

[0098] The second preset time may be one frame time, that is, one sixteenth of a second, which may be comprehensively considered based on the required accuracy and processing speed. However, the interval of the second preset time is (0, 1 second].

[0099] In other equivalent embodiments: the first alarm detector is a surface alarm detector, and the second alarm detector is a linear alarm detector. Or the first alarm detector and the second alarm detector are both dual-technology detectors with passive infrared detection and microwave detection. Or the first alarm detector is a dual-technology detector with passive infrared detection and microwave detection, and the second alarm detector is a forward tilt sensor.

[0100] S7. Summarize the first monitoring result, the second monitoring result and the third monitoring result. If any one of the first monitoring result and the third monitoring result indicates that the vehicle is abnormal or the second monitoring result indicates that a living body is intruded, an abnormal car wash alarm is issued.

[0101] Please refer to Figures 1 to 3 , Embodiment 2 of the present invention is:

[0102] The vehicle movement detection method in the car washing process is based on the above-mentioned embodiment 1, and the second alarm detector is specifically:

[0103] like Figure 2 As shown, a first sensing member 4 and a first rotating shaft 5 are arranged on the front ground of the car wash area, a first motor not shown in the figure is detachably connected to the first rotating shaft 5, and an inclination sensor is arranged on the first sensing member 4. Similarly, a second sensing member 6 and a second rotating shaft 7 are arranged on the rear ground of the car wash area, a second motor not shown in the figure is detachably connected to the second rotating shaft 7, and an inclination sensor is arranged on the first sensing member 4.

[0104] Among them, the first motor is used to control the first rotating shaft 5 to drive the first rotating shaft 5 to rotate from a horizontal state to a forward tilted state and then disconnect from the first rotating shaft 5; the second motor is used to control the second rotating shaft 7 to drive the second sensing element 6 to rotate from a horizontal state to a forward tilted state and then disconnect from the second rotating shaft 7.

[0105] Therefore, in this embodiment, obtaining the third monitoring result of the second alarm detector performing motion detection on the first vehicle 8 in the car wash area in step S3 specifically includes the following steps:

[0106] When the first vehicle 8 in the car wash area is in the vehicle in-place state, the first sensing element 4 located at the front end away from the first vehicle 8 is controlled to rotate from the horizontal state to the forward tilted state tilted toward the front end, and the second sensing element 6 located at the rear end away from the first vehicle 8 is controlled to rotate from the horizontal state to the forward tilted state tilted toward the rear end, until the first sensing element 4 tilts forward on the front end and the second sensing element 6 tilts forward on the rear end;

[0107] A first inclination signal returned by the first inclination sensor located on the first sensing element 4 and a second inclination signal returned by the second inclination sensor located on the second sensing element 6 are acquired in real time, and whether the first vehicle 8 is moving is determined according to changes in the signal values ​​of the first inclination signal or the second inclination signal. If it is moving, a third monitoring result indicating that the vehicle is abnormal is obtained; otherwise, a third monitoring result indicating that the vehicle is normal is obtained.

[0108] Therefore, in this embodiment, if Figure 2 and Figure 3It can be seen that in order to adapt to different vehicle models, the first sensing member 4 and the second sensing member 6 are set at a relatively far position. For vehicles of different lengths, the normal inclination angles are different, and the program needs to determine when to use it as the basic inclination angle. In other equivalent embodiments, when the first sensing member 4, the first rotating shaft 5 and the first motor are in a sliding device, the first motor controls the first sensing member 4 to be at a preset inclination angle and then starts to move until it contacts the vehicle. Similarly, when the second sensing member 6, the second rotating shaft 7 and the second motor are in a sliding device, the second motor controls the second sensing member 6 to be at a preset inclination angle and then starts to move until it contacts the vehicle. In this way, the basic inclination angle is fixed, and no further judgment by the program is required.

[0109] In this embodiment, the above-mentioned motion detection has higher sensitivity, better stability and is easier to implement in a car washing scenario compared to various infrared detections or image detections.

[0110] Please refer to Figure 4 , Embodiment 3 of the present invention is:

[0111] The vehicle movement detection terminal 1 of the car washing process includes a memory 3, a processor 2 and a computer program stored in the memory 3 and executable on the processor 2. When the processor 2 executes the computer program, the steps of the above-mentioned embodiment 1 are implemented.

[0112] In summary, the vehicle movement detection method and terminal for the car wash process provided by the present invention shoots videos of the car wash process when the openable parts of sample vehicles of different vehicle sizes in the car wash area are in a closed state and the unwashed process when the openable parts are in an open state, and then simulates and generates the required image information to train a lower-cost and more accurate image classification model to identify vehicle misoperation. Afterwards, intrusion monitoring and vehicle movement monitoring are performed through the first alarm detector and the second alarm detector, especially through the clever application of the tilt sensor, so as to achieve accurate and effective monitoring and identification of intrusion and vehicle misoperation in the car wash process, which has a low implementation cost and high accuracy.

[0113] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A vehicle movement detection method in a car wash process, characterized in that: The following steps are involved: When the first vehicle arrives at the car wash area, the first sensing element at the front end of the first vehicle is controlled to rotate from a horizontal state to a forward tilted state tilted toward the front end of the vehicle, and the second sensing element at the rear end of the first vehicle is controlled to rotate from a horizontal state to a forward tilted state tilted toward the rear end of the vehicle, until the first sensing element tilts forward on the front end of the vehicle and the second sensing element tilts forward on the rear end of the vehicle; A first inclination signal returned by the first inclination sensor on the first sensing element and a second inclination signal returned by the second inclination sensor on the second sensing element are acquired in real time, and whether the first vehicle is moving is determined according to changes in the signal values ​​of the first inclination signal or the second inclination signal; if it is moving, a third monitoring result indicating that the vehicle is abnormal is obtained; otherwise, a third monitoring result indicating that the vehicle is normal is obtained.

2. The vehicle movement detection method in a car washing process according to claim 1, characterized in that: The first sensing member far from the front end of the first vehicle is respectively controlled to rotate from a horizontal state to a forward tilted state tilted toward the front end, and the second sensing member far from the rear end of the first vehicle is respectively controlled to rotate from a horizontal state to a forward tilted state tilted toward the rear end, until the first sensing member tilts forward on the front end and the second sensing member tilts forward on the rear end, further comprising: Controlling the first sensing element to start moving after being at a preset inclination angle until it contacts the first vehicle, thereby obtaining a first basic inclination angle; Controlling the second sensing element to start moving after being at a preset inclination angle until it contacts the first vehicle, thereby obtaining a second basic inclination angle; A normal inclination angle of the first vehicle is determined based on the first basic inclination angle and the second basic inclination angle.

3. The vehicle movement detection method in a car washing process according to claim 1, characterized in that: The method of respectively controlling a first sensing element at a front end of the first vehicle away from the first vehicle to rotate from a horizontal state to a forward tilted state tilted toward the front end and a second sensing element at a rear end of the first vehicle away from the first vehicle to rotate from a horizontal state to a forward tilted state tilted toward the rear end includes: A second monitoring result of the first alarm detector performing live intrusion detection on the entrance and exit of the car wash area is obtained.

4. The vehicle movement detection method in a car washing process according to claim 3, characterized in that: The first alarm detector is a first camera; Obtaining a real-time monitoring image of the entrance and exit of the car wash area and the car wash area by the first camera, and determining whether two real-time monitoring images separated by a second preset time have differences at the entrance and exit of the car wash area; if so, generating a second monitoring result that a living body intrusion is found; otherwise, generating a second monitoring result that no living body intrusion is found; It is determined whether there is a difference between the two real-time monitoring images separated by a second preset time on the car wash area. If so, a third monitoring result of a normal vehicle is generated; otherwise, a third monitoring result of a normal vehicle is generated.

5. The vehicle movement detection method in a car washing process according to claim 3, characterized in that: The step of obtaining the second monitoring result of the first alarm detector performing live intrusion detection on the entrance and exit of the car wash area includes: Acquire first video information captured by a first camera of sample vehicles of different vehicle sizes located in the car wash area during a car wash process and second video information captured when the vehicles are not washed, wherein the first camera is used to monitor all openable parts of the sample vehicles located in the car wash area, wherein all openable parts of the sample vehicles in the first video information are in a closed state, and the second video information includes video information of each openable part of the sample vehicles being in an open state in turn, wherein the openable parts include windows, sunroofs, doors, and trunks; Obtaining first picture information of each sample vehicle at each car washing step from the first video information, and marking each first picture information as a normal vehicle and then adding it to the training image set; Acquire second picture information of each openable component of each sample vehicle in an open state from the second video information, simulate and obtain third picture information of each openable component of each sample vehicle in an open state during a car wash process based on the first picture information and the second picture information, and mark each piece of the third picture information as a vehicle abnormality and add it to the training image set; An image classification model capable of distinguishing between a normal vehicle and an abnormal vehicle is obtained by training the training image set; A surveillance video of a first vehicle in a car wash process shot in real time by the first camera is obtained, and a frame of surveillance image to be judged is obtained from the surveillance video at every first preset time interval as an input of the trained image classification model to obtain a first monitoring result of the image classification model, wherein the first monitoring result is that the vehicle is normal or the vehicle is abnormal.

6. The vehicle movement detection method in a car washing process according to claim 5, characterized in that: If there is one first camera, the first camera is located in the upper middle of the rear of the vehicle to be washed in the car washing area; If there are two first cameras, the two first cameras are located above both ends of the diagonal line of the vehicle to be cleaned; If there are three first cameras, the three first cameras are respectively located above both sides and the rear side of the vehicle to be cleaned.

7. The vehicle movement detection method in a car washing process according to claim 3, characterized in that: The first alarm detector is a surface-type alarm detector.

8. The vehicle movement detection method in a car washing process according to claim 3, characterized in that: The first alarm detector is a dual technology detector with passive infrared detection and microwave detection.

9. A vehicle movement detection terminal for a car wash process, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the vehicle movement detection method for a car wash process as described in any one of claims 1-8 is implemented.

Images