Vehicle pre-inspection data processing method and platform

Abstract

The present invention provides a vehicle pre-inspection data processing method and platform, which receive a pre-inspection request from a user end, adjust the shooting position of each detection device according to the real-time shooting data of the pre-inspected vehicle in the pre-inspection area by each detection device, and obtain the pre-inspection image collected by the detection device after the adjustment; perform defect detection on the pre-inspected vehicle according to the pre-inspection image to obtain pre-inspection problem data, and determine the defect attributes of the pre-inspection problem data; configure a corresponding defect collection strategy for the pre-inspection problem data based on the defect attributes, and the defect collection strategy includes a detail collection strategy and an overall collection strategy; control the detection device to perform secondary data collection on the pre-inspected vehicle according to the defect collection strategy, and obtain the vehicle pre-inspection data and send it to the management end.

Description

Technical Field

[0001] The present invention relates to image processing technology, and in particular to a vehicle pre-inspection data processing method and platform. Background Art

[0002] Vehicle pre-inspection refers to a series of pre-inspection measures on the vehicle to ensure vehicle safety, performance and reliability, including appearance inspection, mechanical inspection, etc.

[0003] In the prior art, when inspecting the appearance of a vehicle, pre-inspection personnel usually observe the body, lights and other parts directly with the naked eye to check whether there are obvious defects, such as scratches, dents, etc. This method is simple and direct, but is limited by the subjectivity and experience of the pre-inspection personnel, and may result in misjudgments or omissions, thereby reducing the accuracy of vehicle pre-inspection.

[0004] Therefore, how to improve the accuracy of vehicle pre-inspection has become an urgent problem that needs to be solved today. Summary of the invention

[0005] The present invention provides a vehicle pre-inspection data processing method and platform, which can improve the accuracy of vehicle pre-inspection.

[0006] A first aspect of the present invention provides a vehicle pre-inspection data processing method, comprising:

[0007] Receive a pre-inspection request from a user terminal, adjust the shooting position of each detection device according to the real-time shooting data of the pre-inspected vehicle in the pre-inspection area by each detection device, and obtain the pre-inspection image collected by the detection device after adjustment;

[0008] Performing defect detection on the pre-inspected vehicle according to the pre-inspection image to obtain pre-inspection problem data, and determining defect attributes of the pre-inspection problem data;

[0009] Based on the defect attributes, a corresponding defect collection strategy is configured for the pre-check problem data, wherein the defect collection strategy includes a detail collection strategy and an overall collection strategy;

[0010] The detection equipment is controlled according to the defect collection strategy to perform secondary data collection on the pre-inspected vehicle, and the obtained vehicle pre-inspection data is sent to the management end.

[0011] Optionally, in a possible implementation of the first aspect, receiving a pre-inspection request from a user terminal, adjusting a shooting position of each detection device according to real-time shooting data of a pre-inspected vehicle in a pre-inspection area by each detection device, and obtaining a pre-inspection image collected by the detection device after adjustment, including:

[0012] In response to the pre-inspection request, acquiring real-time shooting data of each of the detection devices, the pre-inspection area including a plurality of detection devices;

[0013] Extracting the vehicle contour corresponding to the pre-inspected vehicle in each of the real-time shooting data, and obtaining the contour center point of the vehicle contour;

[0014] Acquire a shooting center point corresponding to a shooting interface of the detection device, and determine a direction from the contour center point to the shooting center point as an adjustment direction;

[0015] Adjusting the shooting position of the detection device based on the adjustment direction until the shooting center point coincides with the contour center point and the adjustment is stopped;

[0016] Determining the shooting ratio corresponding to the pre-inspected vehicle according to the ratio of the contour area of ​​the vehicle contour to the total shooting area corresponding to the detection equipment;

[0017] When the shooting ratio is not in the optimal ratio range, the detection device is zoomed and adjusted in sequence according to multiple preset magnification focal lengths in the magnification list until the shooting ratio is in the optimal ratio range, then the zoom adjustment of the detection device is stopped, and the pre-inspection image collected by the detection device is obtained.

[0018] Optionally, in a possible implementation manner of the first aspect, performing defect detection on the pre-inspection vehicle according to the pre-inspection image to obtain pre-inspection problem data, and determining the defect attribute of the pre-inspection problem data includes:

[0019] Perform defect detection on the pre-inspection image based on a defect analysis model to obtain a defect contour corresponding to the pre-inspection vehicle in the pre-inspection image, and obtain the pre-inspection problem data according to the defect contour;

[0020] Acquire a first length of the defect profile in the vertical direction and a second length of the defect profile in the transverse direction;

[0021] If both the first length and the second length are smaller than the acquisition length threshold, determining that the defect attribute of the defect contour is a detail attribute;

[0022] If the first length and / or the second length is greater than or equal to the acquisition length threshold, determining the defect attribute of the defect profile as an overall attribute;

[0023] Configuring a corresponding defect collection strategy for the pre-check problem data based on the defect attribute includes:

[0024] The detail acquisition strategy is configured for the defect contour of the detail attribute, and the overall acquisition strategy is configured for the defect contour of the overall attribute.

[0025] Optionally, in a possible implementation of the first aspect, controlling the detection device to perform secondary data collection on the pre-inspected vehicle according to the defect collection strategy, and obtaining vehicle pre-inspection data to send to the management end includes:

[0026] Determine a plurality of detail collection positions corresponding to the detection device according to the detail collection strategy, and control the detection device to perform secondary data collection on the defect contour based on the detail collection positions to obtain a type of collection data;

[0027] Based on the overall acquisition strategy, determine that the larger first length or the second length is the defect length, and the smaller first length or the second length is the defect width;

[0028] Determine the vertical direction or the horizontal direction corresponding to the defect length as the acquisition direction, and compare the defect width with a reference defect width;

[0029] When the defect width is smaller than the reference defect width, controlling the detection device to perform fixed data acquisition on the defect profile based on the acquisition direction to obtain second type of acquisition data;

[0030] When the defect width is greater than or equal to the reference defect width, controlling the detection device to perform adaptive data acquisition on the defect profile based on the acquisition direction to obtain two types of acquisition data;

[0031] The vehicle pre-inspection data is obtained according to the first type of collected data and / or the second type of collected data and sent to the management end.

[0032] Optionally, in a possible implementation manner of the first aspect, a plurality of detail collection positions corresponding to the detection device are determined according to the detail collection strategy, and the detection device is controlled to perform secondary data collection on the defect contour based on the detail collection positions to obtain a type of collection data, including:

[0033] Acquire the defect center point of the defect contour, and determine the direction from the image center point of the preview image to the defect center point as the moving direction;

[0034] Determine the point distance between the center point of the defect and the center point of the image, perform actual conversion on the point distance, and obtain the moving distance of the detection device;

[0035] Determine, according to the moving direction and the moving distance, a first detail collection position corresponding to the detection device;

[0036] Controlling the detection device to move to the detail collection position, and acquiring the current defect contour in the shooting interface corresponding to the detection device;

[0037] When the center point of the current defect contour coincides with the shooting center point of the shooting interface, obtaining a type of acquisition image shot by the detection device;

[0038] Determine that the lens shooting direction corresponding to the detection device is the advancing shooting direction, obtain multiple detail acquisition positions in the advancing shooting direction, control the detection device to perform image acquisition based on each of the detail acquisition positions, obtain multiple first-class acquisition images, and obtain a class of acquisition data according to the first-class acquisition images.

[0039] Optionally, in a possible implementation manner of the first aspect, a plurality of detail acquisition positions in the advancing shooting direction are acquired, and the detection device is controlled to perform image acquisition based on each of the detail acquisition positions to obtain a plurality of first-class acquired images, including:

[0040] Acquire the current interval distance between the current position point of the detection device and the vehicle position point in the advancing shooting direction;

[0041] Retrieving a detail shooting correspondence table, traversing a plurality of preset distance intervals in the detail shooting correspondence table based on the current interval distance, and determining a preset shooting quantity corresponding to the preset interval interval where the current interval distance is located as a target shooting quantity;

[0042] Obtain the advancement distance corresponding to the target shooting quantity, and determine the position point at the advancement distance from the current position point in the advancement shooting direction as the next detail collection position;

[0043] Controlling the detection device to move to the next detail acquisition position to acquire an image, and obtaining the next acquired image of the first type;

[0044] Repeat the above steps of determining the detail acquisition position according to the advancement distance, and obtain a type of acquisition image acquired by the detection device at each of the detail acquisition positions, until the number of the type of acquisition images is equal to the target shooting number and stop the above steps.

[0045] Optionally, in a possible implementation manner of the first aspect, when the defect width is smaller than the reference defect width, controlling the detection device to perform fixed data acquisition on the defect profile based on the acquisition direction to obtain two types of acquired data, including:

[0046] When the defect width is smaller than the reference defect width, retrieving the reference optimal distance, and determining the virtual acquisition span of the detection device in the acquisition direction under the reference optimal distance;

[0047] Generating a plurality of division lines perpendicular to the acquisition direction according to the virtual acquisition span, and dividing the defect contour based on the division lines to obtain a plurality of sub-defect contours;

[0048] Acquire the first sub-defect contour in the acquisition direction as the target contour, and determine the initial position point corresponding to the detection device according to the center point of the target contour and the image center point of the pre-inspection image;

[0049] Control the detection device to move to the initial position point, and obtain a first distance between the initial position point and the vehicle position point in the shooting direction of the lens corresponding to the detection device;

[0050] According to the reference optimal distance, the detection device is moved and adjusted in the shooting direction of the lens until the first distance is equal to the reference optimal distance, and a fixed collection point corresponding to the detection device is determined;

[0051] Controlling the detection device to perform image acquisition based on the fixed acquisition points to obtain two types of acquired images;

[0052] Acquire the actual acquisition span corresponding to the virtual acquisition span, determine the position point at the actual acquisition span from the fixed position point in the acquisition direction as the next fixed acquisition point, and control the detection device to go to the next fixed acquisition point to perform image acquisition to obtain the next second-category acquisition image;

[0053] Repeat the above steps of determining fixed acquisition points according to the acquisition direction and obtaining two types of acquisition images according to the fixed acquisition points until the two types of acquisition images corresponding to the sub-defect contours are photographed, and stop the above steps to obtain two types of acquisition data according to the two types of acquisition images.

[0054] Optionally, in a possible implementation manner of the first aspect, when the defect width is greater than or equal to the reference defect width, controlling the detection device to perform adaptive data acquisition on the defect profile based on the acquisition direction to obtain two types of acquired data, including:

[0055] When the defect width is greater than or equal to the reference defect width, determining a plurality of division lines corresponding to the defect contour in the acquisition direction, and dividing the defect contour according to the division lines to obtain a plurality of sub-defect contours;

[0056] Acquire the first sub-defect contour in the acquisition direction as the target contour, and determine the initial position point corresponding to the detection device according to the center point of the target contour and the image center point of the pre-inspection image;

[0057] Control the detection device to move to the initial position point, and obtain a second distance between the initial position point and the vehicle position point in the shooting direction of the lens corresponding to the detection device;

[0058] According to the optimal collection distance corresponding to the sub-defect contour, the detection device is moved and adjusted in the lens shooting direction corresponding to the detection device until the second distance is equal to the optimal collection distance, and a fixed collection point corresponding to the detection device is determined;

[0059] Controlling the detection device to perform image acquisition based on the fixed acquisition points to obtain two types of acquired images;

[0060] Acquire the interval distance between the first sub-defect contour and the next sub-defect contour in the acquisition direction, perform actual conversion on the interval distance to obtain an actual moving distance, and control the detection device to move to the next initial position point based on the actual moving distance and the acquisition direction;

[0061] Repeat the above steps of determining the fixed position point based on the initial position point and obtaining the second type of acquisition image based on the fixed position point until the second type of acquisition image corresponding to each of the sub-defect contours is obtained, and obtain the second type of acquisition data based on the obtained second type of acquisition image.

[0062] Optionally, in a possible implementation manner of the first aspect, when the defect width is greater than or equal to the reference defect width, determining a plurality of division lines corresponding to the defect contour in the acquisition direction, and dividing the defect contour according to the division lines to obtain a plurality of sub-defect contours, including:

[0063] Determine the minimum coordinate value of the defect contour in the acquisition direction, and generate a dividing line corresponding to the minimum coordinate value and perpendicular to the acquisition direction;

[0064] At a virtual acquisition span from the dividing line in the acquisition direction, generating a pending line parallel to the dividing line, dividing the defect contour based on the dividing line and the pending line to obtain a divided contour located between the dividing line and the pending line;

[0065] Acquire the sub-defect width of the divided contour in a direction perpendicular to the acquisition direction, determine the optimal acquisition distance of the sub-defect width, and acquire the current acquisition span of the detection device in the acquisition direction at the optimal acquisition distance;

[0066] Generating the next dividing line parallel to the dividing line at a distance from the current collecting span of the dividing line in the collecting direction;

[0067] Repeat the above step of determining the dividing line until the next dividing line does not intersect with the defect contour, and divide the defect contour according to the dividing line to obtain a plurality of sub-defect contours.

[0068] A second aspect of the present invention provides a vehicle pre-inspection data processing platform, comprising:

[0069] The reservation module is used to receive the pre-inspection request from the user end, adjust the shooting position of each detection device according to the real-time shooting data of the pre-inspected vehicle in the pre-inspection area by each detection device, and obtain the pre-inspection image collected by the detection device after adjustment;

[0070] A detection module, configured to perform defect detection on the pre-inspected vehicle according to the pre-inspection image, obtain pre-inspection problem data, and determine defect attributes of the pre-inspection problem data;

[0071] A configuration module, configured to configure a corresponding defect collection strategy for the pre-check problem data based on the defect attribute, wherein the defect collection strategy includes a detail collection strategy and an overall collection strategy;

[0072] The pre-inspection module is used to control the detection equipment to perform secondary data collection on the pre-inspected vehicle according to the defect collection strategy, and obtain the vehicle pre-inspection data and send it to the management end.

[0073] The beneficial effects of the present invention are as follows:

[0074] The present invention can improve the accuracy of vehicle pre-inspection. When inspecting the appearance of a vehicle, the present invention will collect images of the vehicle through a detection device, and then determine whether the vehicle has defects based on the collected image data, and when the vehicle has defects, the detection device will be controlled to collect secondary images of the defects, thereby improving the accuracy of vehicle appearance inspection, and being able to automatically detect defects, thereby improving the efficiency of vehicle pre-inspection.

[0075] When performing secondary data acquisition on vehicle defect data, the present invention will configure different acquisition methods according to different defect attributes to perform secondary acquisition of defect images. When performing image acquisition on smaller defects, the present invention will use a push-and-move image acquisition method to perform secondary image acquisition on the defects, so that detailed images of the defects can be acquired, thereby improving the accuracy of acquisition. When performing secondary image acquisition on defects with larger spans, the present invention will use a linear-movement image acquisition method to acquire a relatively complete defect image, thereby improving the accuracy of acquisition.

[0076] When pre-inspecting a vehicle, the present invention will also adaptively adjust the position of the detection equipment according to the different vehicles, so that the vehicle outline in the image data collected by the detection equipment can be located in the center position and the proportion can reach the best, thereby improving the accuracy of defect detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 It is a flow chart of a vehicle pre-inspection data processing method provided by an embodiment of the present invention;

[0078] Figure 2 is a schematic diagram of a detail collection strategy provided by an embodiment of the present invention;

[0079] Figure 3 is a schematic diagram of an overall collection strategy provided by an embodiment of the present invention;

[0080] Figure 4 It is a structural schematic diagram of a vehicle pre-inspection data processing platform provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0081] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0082] See also Figure 1 , is a flow chart of a vehicle pre-inspection data processing method provided by an embodiment of the present invention, Figure 1 The execution subject of the method shown may be a software and / or hardware device. The execution subject of the present application may include but is not limited to at least one of the following: user equipment, network equipment, etc. Among them, user equipment may include but is not limited to computers, smart phones, personal digital assistants (PDA) and the electronic devices mentioned above. Network equipment may include but is not limited to a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of computers or network servers, wherein cloud computing is a type of distributed computing, a super virtual computer composed of a group of loosely coupled computers. This embodiment does not limit this. It includes steps S1 to S4, as follows:

[0083] S1, receiving a pre-inspection request from a user terminal, adjusting the shooting position of each detection device according to the real-time shooting data of the pre-inspected vehicle in the pre-inspection area by each detection device, and obtaining the pre-inspection image collected by the adjusted detection device.

[0084] When inspecting the appearance of a vehicle, this solution will capture images of the vehicle through the inspection equipment, and then determine whether the vehicle has any defects based on the collected image data. If the vehicle has defects, the inspection equipment will be controlled to capture a second image of the defects, thereby improving the accuracy of the vehicle appearance inspection, and automatically performing defect detection, thereby improving the efficiency of vehicle pre-inspection.

[0085] Among them, the user end can be a terminal held by a user who needs to perform pre-inspection, the detection device can be a camera, the pre-inspection area refers to the area where vehicles are pre-inspected, the pre-inspection vehicle refers to the vehicle waiting for pre-inspection, the real-time shooting data refers to the real-time shooting screen of the detection device, and the pre-inspection image refers to the image taken of the vehicle.

[0086] In actual applications, multiple detection devices can be set up in the pre-inspection area, so that the vehicle can be fully inspected through the detection devices. For example, a detection device can be configured on the top and four sides of the pre-inspection area, so that the top, front, rear, left and right sides of the vehicle can be imaged from corresponding perspectives, and the vehicle can be fully inspected through the obtained top view, front view, rear view, left view and right view.

[0087] It is understandable that due to different vehicle models, the appearance of the vehicle may also be different. If the shooting position of the detection equipment is kept unchanged, the collected image data may be poor, thereby reducing the accuracy of the detection. Therefore, the detection equipment can adjust the shooting position according to the pre-inspected vehicle so that the position of the detection equipment when collecting the pre-inspection image can be optimal.

[0088] Based on the above embodiment, the specific implementation of step S1 may be:

[0089] S11, in response to the pre-inspection request, obtaining real-time shooting data of each of the detection devices, the pre-inspection area including a plurality of detection devices.

[0090] In actual applications, when a user wants to pre-inspect a vehicle, he can send a corresponding pre-inspection request. After receiving the pre-inspection request, the server can control various detection devices to perform vehicle pre-inspection.

[0091] S12, extracting the vehicle contour corresponding to the pre-inspected vehicle in each of the real-time shooting data, and acquiring the contour center point of the vehicle contour.

[0092] It can be understood that in order to achieve the best effect of the captured image data, the vehicle can be located at the center of the shooting interface during shooting. Specifically, the position of the detection equipment during shooting can be adjusted according to the vehicle outline in the real-time shooting picture, so that the adjusted vehicle outline can be located at the center of the shooting interface, and then the pre-inspected vehicle in the captured image can be located at the center of the image.

[0093] In some embodiments, when obtaining the contour center point, the vehicle contour can be coordinate processed, and then the coordinates corresponding to the contour center point are determined according to the middle values ​​of the maximum and minimum values ​​on the X-axis and Y-axis, and the position point of the contour center point is determined by the coordinates.

[0094] S13, obtaining a shooting center point corresponding to the shooting interface of the detection device, and determining a direction from the contour center point to the shooting center point as an adjustment direction.

[0095] It can be understood that in order to make the contour center point and the shooting center point coincide with each other, the direction from the contour center point to the shooting center point can be determined as the adjustment direction, so that the detection device can be subsequently moved according to the adjustment direction.

[0096] S14, adjusting the shooting position of the detection device based on the adjustment direction, and stopping the adjustment when the shooting center point coincides with the contour center point.

[0097] For example, when the adjustment direction is from left to right, the detection device may move from the original position to the right until the shooting center point coincides with the contour center point and stops moving.

[0098] S15, determining a shooting ratio corresponding to the pre-inspected vehicle according to a ratio of a contour area of ​​the vehicle contour to a total shooting area corresponding to the detection equipment.

[0099] It is understandable that due to different vehicle models, the proportion of different vehicles in the shooting interface may also be different. In order to capture the complete data of the vehicle, the detection device can initially capture the focal length of the largest vehicle, and then subsequently enlarge and adjust the focal length of the detection device based on the shooting proportion, so that the proportion of the vehicle outline in the shooting screen can reach the best. Among them, the total shooting area refers to the screen area of ​​the shooting interface.

[0100] S16, when the shooting ratio is not in the optimal ratio range, the detection device is magnified and zoomed in sequence according to the multiple preset magnification focal lengths in the magnification list until the shooting ratio is in the optimal ratio range, then the zoom adjustment of the detection device is stopped, and the pre-inspection image collected by the detection device is obtained.

[0101] The above-mentioned optimal proportion range refers to the best proportion range of the vehicle outline in the shooting picture. If the shooting proportion is not in the optimal proportion range, it means that the vehicle outline at this time may be too small. Therefore, the image can be enlarged by adjusting the focal length of the lens of the detection equipment.

[0102] The zoom list refers to a list for zooming and adjusting the focal length of the lens of the detection device, which includes multiple pre-set zoom focal length values, namely the preset zoom focal lengths. When the detection device is zoomed, the detection device can be controlled to adjust the focal length in sequence according to the preset zoom focal lengths in the zoom list until the shooting proportion corresponding to the vehicle outline in the picture is within the optimal proportion range, and then the zoom adjustment of the detection device is stopped.

[0103] The position and proportion of the vehicle outline in the pre-inspection image captured in the above manner can be optimized, thereby improving the accuracy of subsequent inspections.

[0104] S2, performing defect detection on the pre-inspection vehicle according to the pre-inspection image to obtain pre-inspection problem data, and determining defect attributes of the pre-inspection problem data.

[0105] After obtaining the pre-inspection image, this solution will use the pre-inspection image to detect defects on the pre-inspected vehicle, so that further image acquisition can be performed on the defects after the defects are detected, so that the management end can view the specific information of the defects and make targeted processing strategies. Among them, the pre-inspection problem data refers to the defect data corresponding to the pre-inspection vehicle.

[0106] When further performing image acquisition on defects in the subsequent process, this solution will configure different acquisition strategies for them according to different defect attributes. Specifically, the above-mentioned pre-inspection problem data and the defect attributes of the pre-inspection problem data can be obtained through the following steps:

[0107] S21, performing defect detection on the preview image based on a defect analysis model to obtain a defect outline corresponding to the preview vehicle in the preview image, and obtaining the preview problem data according to the defect outline.

[0108] The above-mentioned defect analysis model refers to a model for performing defect detection on vehicle image data. In practical applications, the model can be configured with a corresponding defect detection algorithm, such as a deep learning algorithm, so that the model can automatically perform defect detection on the pre-inspection image, identify the defect contour corresponding to the pre-inspected vehicle in the pre-inspection image, and obtain pre-inspection problem data based on the defect contour.

[0109] S22, obtaining a first length of the defect profile in the vertical direction and a second length of the defect profile in the transverse direction.

[0110] When determining the pre-inspection problem data, that is, the defect attributes corresponding to the defect contour, this solution will judge whether it is suitable for the image acquisition method of advancing movement or the image acquisition method of linear movement based on the length of the defect contour in the vertical and horizontal directions, and subsequently configure the corresponding defect attributes for it, and configure the corresponding image acquisition method for it through the defect attributes.

[0111] Among them, the image acquisition method of advancing movement is suitable for smaller defects, such as small dents, etc. In this way, image data corresponding to defects at different distances can be obtained, so that the detailed features of the defects can be displayed in detail.

[0112] The linear motion image acquisition method is suitable for defects with larger spans, such as longer scratches. This method can obtain defect data with higher accuracy and ensure the integrity of the defect data.

[0113] S23: If both the first length and the second length are smaller than a collection length threshold, determining that the defect attribute of the defect contour is a detail attribute.

[0114] If both the first length and the second length are smaller than the acquisition length threshold, it indicates that the defect contour may be small, such as the small concave contour mentioned above, so detail attributes can be configured for it. The detail attributes are acquisition attributes corresponding to the image acquisition method of advancing movement, and the acquisition length threshold is a judgment threshold used to judge the contour span, which can be set in advance.

[0115] S24: If the first length and / or the second length is greater than or equal to the acquisition length threshold, determine that the defect attribute of the defect contour is an overall attribute.

[0116] If the first length and / or the second length is greater than or equal to the acquisition length threshold, it means that the length span of the defect profile in at least one direction exceeds the threshold, such as the long scratch profile mentioned above, so the overall attribute can be configured for it. The overall attribute is the acquisition attribute corresponding to the linear moving image acquisition method.

[0117] S3, configuring a corresponding defect collection strategy for the pre-check problem data based on the defect attribute, wherein the defect collection strategy includes a detail collection strategy and an overall collection strategy.

[0118] Among them, the defect acquisition strategy is the strategy of controlling the detection equipment to perform secondary image acquisition of defects, the detail acquisition strategy is the strategy of controlling the detection equipment to use a push-movement method to acquire images, and the overall acquisition strategy is the strategy of controlling the detection equipment to use a linear movement method to acquire images.

[0119] Specifically, the following steps may be used to configure a corresponding defect collection strategy for the pre-check problem data based on the defect attributes:

[0120] The detail acquisition strategy is configured for the defect contour of the detail attribute, and the overall acquisition strategy is configured for the defect contour of the overall attribute.

[0121] It is understandable that, since the defect attributes and the defect acquisition strategies correspond to each other, the corresponding defect acquisition strategies can be configured for the corresponding defect contours through the above steps, so that secondary image acquisition can be performed on the corresponding defect contours through the defect acquisition strategies in the subsequent process.

[0122] S4, controlling the detection equipment to perform secondary data collection on the pre-inspected vehicle according to the defect collection strategy, and obtaining vehicle pre-inspection data to send to the management end.

[0123] Among them, the vehicle pre-inspection data includes image data of defects collected for the second time. The defect data collected for the second time enables management personnel to view the defects more clearly and accurately, so as to make targeted processing strategies.

[0124] Based on the above embodiment, the specific implementation of step S4 may be:

[0125] S41, determining a plurality of detail collection positions corresponding to the detection device according to the detail collection strategy, and controlling the detection device to perform secondary data collection on the defect contour based on the detail collection positions to obtain a type of collection data.

[0126] When image acquisition is performed by a propulsion movement method, multiple positions for image acquisition in the propulsion direction, i.e., the above-mentioned detail acquisition positions, can be determined, and then the detection equipment can be controlled to move to each detail acquisition position from far to near in turn for secondary image acquisition, thereby obtaining multiple image data acquired according to the propulsion method, i.e., the above-mentioned type of acquisition data.

[0127] In some embodiments, step S41 may be implemented through steps S411 to S416, as follows:

[0128] S411, obtaining the defect center point of the defect contour, and determining the direction from the image center point of the preview image to the defect center point as the moving direction.

[0129] It is understandable that the defect outline is not necessarily located at the center position of the vehicle outline. In order to ensure that the defect outline is located at the center position during the secondary acquisition, the detection equipment can be moved according to the center point of the defect outline, so that the center point of the defect outline can correspond to the center point of the acquired image when the detection equipment performs secondary image acquisition. Specifically, the moving direction of the detection equipment can be determined based on the direction from the image center point of the pre-inspection image to the defect center point.

[0130] S412, determining the point distance between the defect center point and the image center point, and performing actual conversion on the point distance to obtain the moving distance of the detection device.

[0131] In practical applications, when converting the length in the image into the actual length, the conversion ratio may be related to the magnification of the image. This is because when the magnification of the image is different, the proportional relationship between the length of the object in the image and the actual length of the object will also be different. Therefore, after obtaining the point distance, the point distance can be converted accordingly based on the magnification of the pre-inspection image to obtain the actual moving distance of the detection equipment, that is, the above-mentioned moving distance.

[0132] S413: Determine the first detail collection position corresponding to the detection device according to the moving direction and the moving distance.

[0133] When determining multiple detail acquisition positions, this solution first determines the first detail acquisition position for image acquisition based on the moving direction and moving distance, and then determines the remaining detail acquisition positions based on the first detail acquisition position.

[0134] S414, controlling the detection device to move to the detail collection position, and acquiring a current defect outline in a shooting interface corresponding to the detection device.

[0135] It is understandable that in order to make the defect outline be located at the center during shooting, the lens position may be adjusted according to the current defect outline in the shooting interface, and the current defect outline is the defect outline currently being secondarily captured.

[0136] S415, when the center point of the current defect contour coincides with the shooting center point of the shooting interface, a type of acquisition image shot by the detection device is obtained.

[0137] Similarly, in order to make the acquired defect contour be located at the center of the shooting interface, the image can be shot again when the shooting center point coincides with the center point of the current defect contour to obtain a type of acquired image.

[0138] S416, determining that the lens shooting direction corresponding to the detection device is the advancing shooting direction, obtaining multiple detail acquisition positions in the advancing shooting direction, controlling the detection device to perform image acquisition based on each of the detail acquisition positions, obtaining multiple first-class acquisition images, and obtaining a class of acquisition data according to the first-class acquisition images.

[0139] See also Figure 2, which is a schematic diagram of a detail collection strategy provided by an embodiment of the present invention. After controlling the detection device to move to the first detail collection position, this solution will determine the direction in which the detection device shoots the pre-inspection vehicle as the advancing shooting direction, and then determine multiple detail collection positions from far to near in the advancing shooting direction, so that secondary image collection of the defect contour from far to near can be achieved, and image data corresponding to the defect contour at different distances can be obtained, and then multiple detail images of the defect contour can be collected, thereby improving the accuracy of the pre-inspection.

[0140] Based on the above embodiment, the specific implementation of "obtaining multiple detail acquisition positions in the advancing shooting direction, controlling the detection device to perform image acquisition based on each of the detail acquisition positions, and obtaining multiple first-class acquisition images" in step S416 can be:

[0141] S4161, obtaining the current interval distance between the current position point of the detection device and the vehicle position point in the advancing shooting direction.

[0142] In actual applications, the vehicle position point can be the parking position point of the pre-inspection vehicle during pre-inspection. The position point can be set in advance. The current interval distance between the detection equipment and the pre-inspection vehicle in the advancing shooting direction can be determined through the position point and the current position point of the detection equipment. Therefore, the interval distance of the detection equipment when performing advancing shooting can be determined based on the current interval distance.

[0143] S4162, retrieve a detail shooting correspondence table, traverse a plurality of preset distance intervals in the detail shooting correspondence table based on the current interval distance, and determine that a preset shooting quantity corresponding to the preset interval interval where the current interval distance is located is a target shooting quantity.

[0144] Among them, the detail shooting correspondence table refers to a correspondence table for determining the number of shots and shooting distances when the detection equipment is shooting in a propulsion mode, including multiple intervals corresponding to shooting distances, namely the above-mentioned preset distance intervals, and the number of image shots corresponding to each interval, namely the above-mentioned preset shooting number.

[0145] It is understandable that the shooting distance between the detection equipment and the pre-inspected vehicle is different, and the number of images that can be captured may also be different. For example, the farther the distance, the more images that may be captured. Therefore, the number of images that the detection equipment can take can be determined based on the current interval distance, that is, the above-mentioned target shooting number.

[0146] S4163, obtaining the advancing distance corresponding to the target shooting quantity, and determining a position point at the advancing distance from the current position point in the advancing shooting direction as the next detail collection position.

[0147] In practical applications, different numbers of image capture can be configured in advance with the corresponding interval distances during the push-forward shooting, that is, the above-mentioned push-forward distance, so that the next detail collection position in the push-forward shooting direction can be determined according to the push-forward distance.

[0148] S4164, controlling the detection device to move to the next detail acquisition position to acquire an image, and obtaining the next acquired image of the first type.

[0149] It is worth mentioning that when the inspection equipment collects images at various detail collection positions, the pictures are taken only when the defect outline is located at the center of the shooting interface.

[0150] S4165, repeat the above steps of determining the detail collection position according to the advancing distance, and obtain a type of collection images collected by the detection device at each of the detail collection positions, and stop the above steps when the number of the type of collection images is equal to the target shooting number.

[0151] After obtaining the next detail collection position, the detail collection position can be used as a reference to determine the next detail collection position at the advancing distance in the advancing shooting direction, and the image is captured when the center point of the defect contour coincides with the center point of the shooting interface, and so on, until the number of collected images of one type is consistent with the target shooting number, and the image collection step is stopped.

[0152] S42: Determine, based on the overall acquisition strategy, that the larger first length or the second length is the defect length, and the smaller first length or the second length is the defect width.

[0153] When performing secondary image acquisition of the defect contour according to the overall acquisition strategy, this solution will first determine the direction of the detection equipment during mobile acquisition, and then perform image acquisition of the defect contour according to the direction, so as to obtain complete image data corresponding to the defect contour.

[0154] Specifically, the defect length and defect width can be determined first, so that the direction of the detection device when performing secondary image acquisition can be determined according to the defect length, and the distance between the detection device and the pre-inspected vehicle when performing secondary image acquisition can be determined according to the defect width.

[0155] S43, determining the vertical direction or the horizontal direction corresponding to the defect length as the acquisition direction, and comparing the defect width with a reference defect width.

[0156] It can be understood that the defect length corresponds to the larger length span in the defect contour. In order to completely collect the defect contour, the direction corresponding to the defect length can be determined as the collection direction. The defect width is compared with the benchmark defect width in order to determine the shooting distance between the detection equipment and the pre-inspection vehicle when it moves in the collection direction.

[0157] See also Figure 3 , which is a schematic diagram of an overall acquisition strategy provided by an embodiment of the present invention. As can be seen from the figure, when the acquisition direction is from left to right, this solution will control the detection device to take multiple images of the defect contour from left to right, so that a complete defect contour image can be obtained through these images.

[0158] S44, when the defect width is smaller than the reference defect width, controlling the detection device to perform fixed data acquisition on the defect profile based on the acquisition direction to obtain second type of acquisition data.

[0159] The reference defect width refers to the defect width threshold when the detection device performs image acquisition at a configured shooting distance. When the defect width is smaller than the reference defect width, it indicates that the detection device is able to completely acquire the image data corresponding to the defect contour at the configured shooting distance, and there will be no data missing due to the defect contour being too wide. Therefore, the detection device can be controlled to perform fixed data acquisition on the defect contour according to the acquisition direction, that is, to perform data acquisition at a pre-configured shooting distance, and obtain multiple acquired image data, that is, the above-mentioned two types of acquired data.

[0160] In some embodiments, step S44 may be implemented through steps S441 to S448, as follows:

[0161] S441, when the defect width is smaller than the reference defect width, retrieve the reference optimal distance, and determine the virtual acquisition span of the detection device in the acquisition direction at the reference optimal distance.

[0162] The reference optimal distance refers to the shooting distance between the pre-configured detection equipment and the pre-inspected vehicle. It is understandable that the shooting range corresponding to the detection equipment may be different at different shooting distances. In order to determine the multiple position points when the detection equipment moves and shoots, the shooting range corresponding to the detection equipment can be determined based on the reference optimal distance, and then the defect contour is segmented according to the shooting range, and then the detection equipment is controlled to sequentially capture the corresponding images of the multiple sub-defect contours after segmentation.

[0163] The virtual acquisition span refers to the length span after the actual acquisition span of the detection device is virtually converted, and the span corresponds to the length span in the acquisition direction.

[0164] S442: Generate a plurality of division lines perpendicular to the acquisition direction according to the virtual acquisition span, and divide the defect contour based on the division lines to obtain a plurality of sub-defect contours.

[0165] In some embodiments, when dividing the defect contour according to the virtual acquisition span and the acquisition direction, the minimum value of the defect contour in the acquisition direction can be determined first, and then a division line perpendicular to the acquisition direction is generated based on the minimum value, and then the next parallel division line is generated at a distance of the virtual acquisition span from the division line in the acquisition direction, and so on and so forth to generate multiple division lines with equal intervals. The defect contour is divided by multiple division lines to obtain multiple sub-defect contours.

[0166] S443, acquiring the first sub-defect contour in the acquisition direction as a target contour, and determining an initial position point corresponding to the detection device according to a center point of the target contour and an image center point of the preview image.

[0167] Specifically, when determining the initial position point, the moving direction of the detection device can be determined based on the direction from the center point of the image to the center point of the target contour, and the moving distance of the detection device after actual conversion can be determined based on the point distance between the two center points. The initial position point is determined by the moving distance and the moving direction.

[0168] S444, controlling the detection device to move to the initial position point, and obtaining a first distance between the initial position point and the vehicle position point in the shooting direction of the lens corresponding to the detection device.

[0169] It can be understood that after the detection device moves to the initial position point, in order to make the shooting distance between the detection device and the pre-inspection vehicle correspond to the benchmark optimal distance, the current shooting distance between the detection device and the pre-inspection vehicle, that is, the above-mentioned first distance, can also be determined based on the initial position point and the vehicle position point.

[0170] S445, according to the reference optimal distance, the detection device is moved and adjusted in the shooting direction of the lens until the first distance is equal to the reference optimal distance, and a fixed collection point corresponding to the detection device is determined.

[0171] Since the reference optimal distance refers to the shooting distance between the detection device and the pre-inspected vehicle, the detection device can be controlled to move in the lens shooting direction to determine the final position point for image acquisition, that is, the above-mentioned fixed acquisition point.

[0172] S446, controlling the detection device to perform image acquisition based on the fixed acquisition points to obtain two types of acquired images.

[0173] The two types of captured images obtained by the above shooting are the secondary captured images corresponding to the target contour.

[0174] S447, obtaining the actual acquisition span corresponding to the virtual acquisition span, determining the position point at the actual acquisition span from the fixed position point in the acquisition direction as the next fixed acquisition point, and controlling the detection device to go to the next fixed acquisition point to perform image acquisition to obtain the next second-category acquisition image.

[0175] It can be understood that since the defect contour is divided according to the virtual acquisition span, when determining the next fixed position point, the actual acquisition span corresponding to the virtual acquisition span can be determined first, and then the position point in the acquisition direction that is different from the current position point of the detection equipment by the actual acquisition span is determined as the next fixed acquisition point, and the next Class II acquisition image is taken at the next fixed acquisition point.

[0176] S448, repeat the above steps of determining fixed acquisition points according to the acquisition direction and obtaining two types of acquisition images according to the fixed acquisition points, until the two types of acquisition images corresponding to the sub-defect contours are photographed, stop the above steps, and obtain two types of acquisition data according to the two types of acquisition images.

[0177] Specifically, the current fixed acquisition point of the detection device can be used as a reference to continue to determine the next fixed acquisition point at the actual acquisition span in the acquisition direction until the secondary image acquisition operation of the detection device is stopped when the image data corresponding to all sub-defect contours are acquired.

[0178] S45, when the defect width is greater than or equal to the reference defect width, controlling the detection device to perform adaptive data acquisition on the defect contour based on the acquisition direction to obtain two types of acquisition data.

[0179] When the defect width is greater than or equal to the baseline defect width, it means that the detection device is unable to completely collect the image data corresponding to the defect contour according to the configured shooting distance, and data loss may occur due to the defect contour being too wide. Therefore, the defect contour can be adaptively segmented according to the defect width, and then the detection device is controlled to collect images of each segmented sub-contour according to the collection direction to obtain two types of collected data.

[0180] Specifically, step S45 includes steps S451 to S457, which are as follows:

[0181] S451, when the defect width is greater than or equal to the reference defect width, determining a plurality of division lines corresponding to the defect contour in the acquisition direction, and dividing the defect contour according to the division lines to obtain a plurality of sub-defect contours.

[0182] When the defect width is greater than or equal to the reference defect width, this solution will adaptively segment the defect contour according to the defect width to obtain multiple sub-defect contours, so that secondary image acquisition can be performed on each sub-defect contour subsequently.

[0183] In some embodiments, the defect contour may be adaptively segmented by the following steps:

[0184] S4511, determining the minimum coordinate value of the defect contour in the acquisition direction, and generating a dividing line corresponding to the minimum coordinate value and perpendicular to the acquisition direction.

[0185] The dividing line generated according to the minimum coordinate value is the first dividing line of the defect contour in the acquisition direction.

[0186] S4512, generating a pending line parallel to the dividing line at a virtual acquisition span from the dividing line in the acquisition direction, dividing the defect contour based on the dividing line and the pending line to obtain a divided contour located between the dividing line and the pending line.

[0187] It is understandable that when adaptively dividing the defect contour, the defect contour may be divided according to the span corresponding to the pre-configured shooting distance first, and then the shooting distance during acquisition may be adjusted based on the divided contour.

[0188] S4513, obtaining the sub-defect width of the divided outline in a direction perpendicular to the acquisition direction, determining an optimal acquisition distance of the sub-defect width, and obtaining a current acquisition span of the detection device in the acquisition direction at the optimal acquisition distance.

[0189] In practical applications, a table corresponding to defect width intervals and shooting distances can be configured in advance. Different defect width intervals can correspond to different shooting distances, so that the optimal acquisition distance corresponding to the sub-defect width can be determined through the table, that is, the optimal shooting distance. Then, the acquisition range of the detection equipment under the shooting distance can be determined, and the acquisition span of the detection equipment in the acquisition direction can be determined through the acquisition range. The current acquisition span is a virtual length span converted from the actual acquisition span of the detection equipment in the acquisition direction.

[0190] S4514: Generate the next dividing line parallel to the dividing line at a distance from the current collecting span of the dividing line in the collecting direction.

[0191] It is understandable that in order to make the acquired image data better, a corresponding determination can be made according to the current acquisition span when generating the next dividing line.

[0192] S4515, repeat the above step of determining the dividing line until the next dividing line does not intersect with the defect contour, and divide the defect contour according to the dividing line to obtain a plurality of sub-defect contours.

[0193] When obtaining the next dividing line, the current dividing line can be used as a reference to continue to determine the next pending line according to the virtual acquisition span, and determine the dividing contour between the current dividing line and the next pending line. According to the dividing contour, the corresponding current acquisition span is determined again, and the next dividing line is generated, and so on, until the generated dividing line does not intersect with the defect contour and stops generating the dividing line.

[0194] It is worth mentioning that if the maximum length of the sub-defect contour in the direction perpendicular to the acquisition direction is greater than the sub-defect width, the shooting distance for acquiring the sub-defect contour is determined according to the shooting distance corresponding to its maximum length, but the span during mobile acquisition still corresponds to the current acquisition span corresponding to the sub-defect width, thereby ensuring the integrity of data acquisition.

[0195] S452, acquiring the first sub-defect contour in the acquisition direction as the target contour, and determining the initial position point corresponding to the detection device according to the center point of the target contour and the image center point of the pre-inspection image.

[0196] The above method of determining the initial position point is similar to the method of determining the initial position point in step S443 (obtaining the first sub-defect contour in the acquisition direction as the target contour, and determining the initial position point corresponding to the detection device according to the center point of the target contour and the image center point of the pre-inspection image), and will not be repeated here.

[0197] S453, controlling the detection device to move to the initial position point, and obtaining a second distance between the initial position point and the vehicle position point in the shooting direction of the lens corresponding to the detection device.

[0198] The second distance refers to the current shooting distance between the detection device and the pre-inspected vehicle.

[0199] S454, according to the optimal collection distance corresponding to the sub-defect contour, the detection device is moved and adjusted in the lens shooting direction corresponding to the detection device until the second distance is equal to the optimal collection distance, and a fixed collection point corresponding to the detection device is determined.

[0200] It is worth mentioning that if the maximum length of the sub-defect contour in the direction perpendicular to the acquisition direction is greater than the sub-defect width, then when the detection equipment is moved and adjusted, the shooting distance of the detection equipment in the shooting direction of the lens can be made to correspond to the shooting distance corresponding to the maximum length, thereby ensuring the integrity of data collection.

[0201] S455, controlling the detection device to perform image acquisition based on the fixed acquisition points to obtain two types of acquired images.

[0202] The above two types of collected images are captured images corresponding to the target contour.

[0203] S456, obtaining the interval distance between the first sub-defect contour and the next sub-defect contour in the acquisition direction, actually converting the interval distance to obtain an actual moving distance, and controlling the detection device to move to the next initial position point based on the actual moving distance and the acquisition direction.

[0204] When obtaining the interval distance, it can be determined based on the point distance between the center point of the first sub-defect contour and the next sub-defect contour in the acquisition direction. The actual conversion of the interval distance can determine the actual movement distance of the detection device, thereby controlling the detection device to move to the next initial position point based on the actual movement distance and the acquisition direction, and after moving to the initial position point, the detection device is moved again according to the shooting distance corresponding to the corresponding sub-defect contour, so as to determine the fixed position point for final image acquisition.

[0205] S457, repeat the above steps of determining the fixed position point based on the initial position point, and obtaining the second type of acquisition image based on the fixed position point, until the second type of acquisition image corresponding to each of the sub-defect contours is obtained, and obtain the second type of acquisition data based on the obtained second type of acquisition image.

[0206] Specifically, the optimal acquisition distance corresponding to the corresponding sub-defect contour can be determined, and then the detection equipment can be controlled to move in the lens shooting direction until it moves to the position corresponding to the optimal acquisition distance to determine the final fixed position point for image acquisition, and the image is captured at the fixed position point to obtain the second type of acquired image, and so on, finally obtaining the second type of acquired images corresponding to all sub-defect contours.

[0207] S46, obtaining vehicle pre-inspection data according to the first type of collected data and / or the second type of collected data and sending it to the management end.

[0208] Through the above method, the defect data can be collected for the second time to improve the accuracy of defect detection.

[0209] See also Figure 4 , is a schematic diagram of the structure of a vehicle pre-inspection data processing platform provided by an embodiment of the present invention, the vehicle pre-inspection data processing platform includes:

[0210] The reservation module is used to receive the pre-inspection request from the user end, adjust the shooting position of each detection device according to the real-time shooting data of the pre-inspected vehicle in the pre-inspection area by each detection device, and obtain the pre-inspection image collected by the detection device after adjustment;

[0211] A detection module, configured to perform defect detection on the pre-inspected vehicle according to the pre-inspection image, obtain pre-inspection problem data, and determine defect attributes of the pre-inspection problem data;

[0212] A configuration module, configured to configure a corresponding defect collection strategy for the pre-check problem data based on the defect attribute, wherein the defect collection strategy includes a detail collection strategy and an overall collection strategy;

[0213] The pre-inspection module is used to control the detection equipment to perform secondary data collection on the pre-inspected vehicle according to the defect collection strategy, and obtain the vehicle pre-inspection data and send it to the management end.

[0214] Figure 4 The apparatus of the embodiment shown can be used to perform Figure 1 The implementation principles and technical effects of the steps in the method embodiment shown are similar and will not be repeated here.

[0215] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A vehicle pre-inspection data processing method, characterized in that: include: Receive a pre-inspection request from a user terminal, adjust the shooting position of each detection device according to the real-time shooting data of the pre-inspected vehicle in the pre-inspection area by each detection device, and obtain the pre-inspection image collected by the detection device after adjustment; Performing defect detection on the pre-inspected vehicle according to the pre-inspection image to obtain pre-inspection problem data, and determining defect attributes of the pre-inspection problem data; Based on the defect attributes, a corresponding defect collection strategy is configured for the pre-check problem data, wherein the defect collection strategy includes a detail collection strategy and an overall collection strategy; According to the defect collection strategy, the detection equipment is controlled to collect secondary data on the pre-inspected vehicle, and the obtained vehicle pre-inspection data is sent to the management end; Performing defect detection on the pre-inspected vehicle according to the pre-inspection image to obtain pre-inspection problem data, and determining defect attributes of the pre-inspection problem data, including: Perform defect detection on the pre-inspection image based on a defect analysis model to obtain a defect contour corresponding to the pre-inspection vehicle in the pre-inspection image, and obtain the pre-inspection problem data according to the defect contour; Acquire a first length of the defect profile in a vertical direction and a second length of the defect profile in a transverse direction; If both the first length and the second length are smaller than the acquisition length threshold, determining that the defect attribute of the defect contour is a detail attribute; If the first length and / or the second length is greater than or equal to the acquisition length threshold, determining the defect attribute of the defect profile as an overall attribute; Configuring a corresponding defect collection strategy for the pre-check problem data based on the defect attribute includes: configuring the detail acquisition strategy for the defect contour of the detail attribute, and configuring the overall acquisition strategy for the defect contour of the overall attribute; According to the defect collection strategy, the detection device is controlled to perform secondary data collection on the pre-inspected vehicle, and the obtained vehicle pre-inspection data is sent to the management end, including: Determine a plurality of detail collection positions corresponding to the detection device according to the detail collection strategy, and control the detection device to perform secondary data collection on the defect contour based on the detail collection positions to obtain a type of collection data; Based on the overall acquisition strategy, determine that the larger first length or the second length is the defect length, and the smaller first length or the second length is the defect width; Determine the vertical direction or the horizontal direction corresponding to the defect length as the acquisition direction, and compare the defect width with a reference defect width; When the defect width is smaller than the reference defect width, controlling the detection device to perform fixed data acquisition on the defect profile based on the acquisition direction to obtain second type of acquisition data; When the defect width is greater than or equal to the reference defect width, controlling the detection device to perform adaptive data acquisition on the defect profile based on the acquisition direction to obtain two types of acquisition data; The vehicle pre-inspection data is obtained according to the first type of collected data and / or the second type of collected data and sent to the management end; Determine a plurality of detail collection positions corresponding to the detection device according to the detail collection strategy, control the detection device to perform secondary data collection on the defect contour based on the detail collection positions, and obtain a type of collection data, including: Acquire the defect center point of the defect contour, and determine the direction from the image center point of the preview image to the defect center point as the moving direction; Determine the point distance between the center point of the defect and the center point of the image, perform actual conversion on the point distance, and obtain the moving distance of the detection device; Determine, according to the moving direction and the moving distance, a first detail collection position corresponding to the detection device; Controlling the detection device to move to the detail collection position, and acquiring the current defect contour in the shooting interface corresponding to the detection device; When the center point of the current defect contour coincides with the shooting center point of the shooting interface, obtaining a type of acquisition image shot by the detection device; Determine that the lens shooting direction corresponding to the detection device is the advancing shooting direction, obtain multiple detail acquisition positions in the advancing shooting direction, control the detection device to perform image acquisition based on each of the detail acquisition positions to obtain multiple first-class acquisition images, and obtain first-class acquisition data according to the first-class acquisition images; When the defect width is greater than or equal to the reference defect width, the detection device is controlled to perform adaptive data acquisition on the defect profile based on the acquisition direction to obtain two types of acquisition data, including: When the defect width is greater than or equal to the reference defect width, determining a plurality of division lines corresponding to the defect contour in the acquisition direction, and dividing the defect contour according to the division lines to obtain a plurality of sub-defect contours; Acquire the first sub-defect contour in the acquisition direction as the target contour, and determine the initial position point corresponding to the detection device according to the center point of the target contour and the image center point of the pre-inspection image; Control the detection device to move to the initial position point, and obtain a second distance between the initial position point and the vehicle position point in the shooting direction of the lens corresponding to the detection device; According to the optimal collection distance corresponding to the sub-defect contour, the detection device is moved and adjusted in the lens shooting direction corresponding to the detection device until the second distance is equal to the optimal collection distance, and a fixed collection point corresponding to the detection device is determined; Controlling the detection device to perform image acquisition based on the fixed acquisition points to obtain two types of acquired images; Acquire the interval distance between the first sub-defect contour and the next sub-defect contour in the acquisition direction, perform actual conversion on the interval distance to obtain an actual moving distance, and control the detection device to move to the next initial position point based on the actual moving distance and the acquisition direction; Repeat the above steps of determining the fixed acquisition point according to the initial position point and obtaining the second type of acquisition image according to the fixed acquisition point until the second type of acquisition image corresponding to each of the sub-defect contours is obtained, and obtain the second type of acquisition data according to the obtained second type of acquisition image.

2. The method according to claim 1, characterized in that Receiving a pre-inspection request from a user terminal, adjusting the shooting position of each detection device according to the real-time shooting data of the pre-inspected vehicle in the pre-inspection area by each detection device, and obtaining the pre-inspection image collected by the detection device after adjustment, including: In response to the pre-inspection request, acquiring real-time shooting data of each of the detection devices, the pre-inspection area including a plurality of detection devices; Extracting the vehicle contour corresponding to the pre-inspected vehicle in each of the real-time shooting data, and obtaining the contour center point of the vehicle contour; Acquire a shooting center point corresponding to a shooting interface of the detection device, and determine a direction from the contour center point to the shooting center point as an adjustment direction; Adjusting the shooting position of the detection device based on the adjustment direction until the shooting center point coincides with the contour center point and the adjustment is stopped; Determining the shooting ratio corresponding to the pre-inspected vehicle according to the ratio of the contour area of ​​the vehicle contour to the total shooting area corresponding to the detection equipment; When the shooting ratio is not in the optimal ratio range, the detection device is zoomed and adjusted in sequence according to multiple preset magnification focal lengths in the magnification list until the shooting ratio is in the optimal ratio range, then the zoom adjustment of the detection device is stopped, and the pre-inspection image collected by the detection device is obtained.

3. The method according to claim 1, characterized in that Acquiring a plurality of detail acquisition positions in the advancing shooting direction, and controlling the detection device to perform image acquisition based on each of the detail acquisition positions to obtain a plurality of first-class acquisition images, including: Acquire the current interval distance between the current position point of the detection device and the vehicle position point in the advancing shooting direction; Retrieving a detail shooting correspondence table, traversing a plurality of preset distance intervals in the detail shooting correspondence table based on the current interval distance, and determining a preset shooting quantity corresponding to the preset interval interval where the current interval distance is located as a target shooting quantity; Obtain the advancement distance corresponding to the target shooting quantity, and determine the position point at the advancement distance from the current position point in the advancement shooting direction as the next detail collection position; Controlling the detection device to move to the next detail acquisition position to acquire an image, and obtaining the next acquired image of the first type; Repeat the above steps of determining the detail acquisition position according to the advancement distance, and obtain a type of acquisition image acquired by the detection device at each of the detail acquisition positions, until the number of the type of acquisition images is equal to the target shooting number and stop the above steps.

4. The method according to claim 1, characterized in that: When the defect width is smaller than the reference defect width, the detection device is controlled to perform fixed data acquisition on the defect profile based on the acquisition direction to obtain two types of acquisition data, including: When the defect width is smaller than the reference defect width, retrieving the reference optimal distance, and determining the virtual acquisition span of the detection device in the acquisition direction under the reference optimal distance; Generating a plurality of division lines perpendicular to the acquisition direction according to the virtual acquisition span, and dividing the defect contour based on the division lines to obtain a plurality of sub-defect contours; Acquire the first sub-defect contour in the acquisition direction as the target contour, and determine the initial position point corresponding to the detection device according to the center point of the target contour and the image center point of the pre-inspection image; Control the detection device to move to the initial position point, and obtain a first distance between the initial position point and the vehicle position point in the shooting direction of the lens corresponding to the detection device; According to the reference optimal distance, the detection device is moved and adjusted in the shooting direction of the lens until the first distance is equal to the reference optimal distance, and a fixed collection point corresponding to the detection device is determined; Controlling the detection device to perform image acquisition based on the fixed acquisition points to obtain two types of acquired images; Acquire the actual acquisition span corresponding to the virtual acquisition span, determine the position point at the actual acquisition span from the fixed acquisition point in the acquisition direction as the next fixed acquisition point, and control the detection device to go to the next fixed acquisition point to perform image acquisition to obtain the next second-category acquisition image; Repeat the above steps of determining fixed acquisition points according to the acquisition direction and obtaining two types of acquisition images according to the fixed acquisition points until the two types of acquisition images corresponding to the sub-defect contours are photographed, and stop the above steps to obtain two types of acquisition data according to the two types of acquisition images.

5. The method according to claim 1, characterized in that When the defect width is greater than or equal to the reference defect width, a plurality of dividing lines corresponding to the defect contour in the acquisition direction are determined, and the defect contour is divided according to the dividing lines to obtain a plurality of sub-defect contours, including: Determine the minimum coordinate value of the defect contour in the acquisition direction, and generate a dividing line corresponding to the minimum coordinate value and perpendicular to the acquisition direction; At a virtual acquisition span from the dividing line in the acquisition direction, generating a pending line parallel to the dividing line, dividing the defect contour based on the dividing line and the pending line to obtain a divided contour located between the dividing line and the pending line; Acquire the sub-defect width of the divided contour in a direction perpendicular to the acquisition direction, determine the optimal acquisition distance of the sub-defect width, and acquire the current acquisition span of the detection device in the acquisition direction at the optimal acquisition distance; Generating the next dividing line parallel to the dividing line at a distance from the current collecting span of the dividing line in the collecting direction; Repeat the above step of determining the dividing line until the next dividing line does not intersect with the defect contour, and divide the defect contour according to the dividing line to obtain a plurality of sub-defect contours.

6. A processing platform suitable for the vehicle pre-inspection data processing method according to claim 1, characterized in that: include: The reservation module is used to receive the pre-inspection request from the user end, adjust the shooting position of each detection device according to the real-time shooting data of the pre-inspected vehicle in the pre-inspection area by each detection device, and obtain the pre-inspection image collected by the detection device after adjustment; A detection module, configured to perform defect detection on the pre-inspected vehicle according to the pre-inspection image, obtain pre-inspection problem data, and determine defect attributes of the pre-inspection problem data; A configuration module, configured to configure a corresponding defect collection strategy for the pre-check problem data based on the defect attribute, wherein the defect collection strategy includes a detail collection strategy and an overall collection strategy; The pre-inspection module is used to control the detection equipment to perform secondary data collection on the pre-inspected vehicle according to the defect collection strategy, and obtain the vehicle pre-inspection data and send it to the management end.

Images