An artificial intelligence-based automobile outer dimension rapid screening system and method

Abstract

The application discloses a kind of based on artificial intelligence's automobile external dimension rapid screening system and method, belong to automobile detection technical field, its system includes image acquisition module, artificial intelligence identification module, size calculation module, space generation module and screening judgment module. By artificial intelligence automatic identification vehicle type and calculate automobile external dimension, utilize deep learning model and image processing algorithm, more accurately identify vehicle type and measure automobile external dimension, reduce the measurement error caused by human factor, simultaneously can according to preset standard carry out the screening of vehicle external dimension, ensure the legality and effectiveness of detection result, effectively solve the low precision of existing automobile external dimension detection, the problem of low efficiency. The whole system can complete the screening of vehicle external dimension in short time, is suitable for the scene needing rapid detection such as automobile production assembly line, motor vehicle annual inspection station.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of automobile detection, and particularly relates to a vehicle external dimension rapid screening system and method based on artificial intelligence. BACKGROUND

[0002] In the links of automobile production, sales, annual inspection, etc., the detection of the external dimension of the automobile is an important step to ensure that the vehicle meets the relevant standards. The traditional vehicle external dimension detection method usually needs to manually use measuring tools to measure the length, width and height of the vehicle one by one. This method is not only inefficient, but also easily affected by human factors, resulting in inaccurate measurement results and time-consuming and laborious.

[0003] For the motor vehicle annual inspection scene, since the vehicle external dimension is often measured dynamically (i.e., the vehicle drives through the measurement area at a specified speed, and the relevant radar or light curtain type device scans the vehicle to obtain the dimension data), the error is greatly increased, and since the data of unqualified vehicles needs to be constantly rechecked to meet the requirements of the relevant national standards. The above leads to a huge waste of manpower and time cost. SUMMARY

[0004] The purpose of the present application is to provide a vehicle external dimension rapid screening system and method based on artificial intelligence, which solves the problems of low accuracy and low efficiency of existing vehicle external dimension detection.

[0005] The technical scheme adopted by the present application is as follows:

[0006] A vehicle external dimension rapid screening system based on artificial intelligence, comprising an image acquisition module, an artificial intelligence recognition module, a dimension calculation module, a space generation module and a screening judgment module;

[0007] The image acquisition module is used to acquire the appearance image of the measured vehicle and the relative position information between the measured vehicle and the image acquisition module. The image acquisition module comprises a high-definition camera and a laser ranging unit. The high-definition camera is used to shoot the appearance image of the vehicle from two angles of side view and overhead view. The laser ranging unit is used to measure the straight-line distance between the measured vehicle and the image acquisition module.

[0008] The artificial intelligence recognition module is in communication connection with the image acquisition module, and is used to receive the vehicle image and the relative position information acquired by the image acquisition module, identify the vehicle type according to the acquired vehicle image and the relative position information through a deep learning model, and acquire the external dimension allowable range of the corresponding vehicle type according to a preset standard.

[0009] The size calculation module is in communication connection with the image acquisition module, is used for receiving the vehicle image and relative position information acquired by the image acquisition module, and calculating the actual overall size of the vehicle according to the acquired vehicle image and relative position information;

[0010] The space generation module is in communication connection with the artificial intelligence recognition module, and is used for generating a two-dimensional Hui-shaped space model according to the vehicle overall size allowable range, wherein the inner interface and the outer interface of the two-dimensional Hui-shaped space model correspond to the lower limit value and the upper limit value of the vehicle overall size respectively.

[0011] The screening judgment module is in communication connection with the size calculation module and the space generation module, and is used for comparing and analyzing the actual overall size of the vehicle output by the size calculation module with the two-dimensional Hui-shaped space model generated by the space generation module, to judge whether the actual overall size of the vehicle conforms to the overall size allowable range.

[0012] Further, the high-definition camera and the laser ranging unit of the image acquisition module adopt an integrated design, and the laser ranging unit and the lens part in the high-definition camera are located at the same position.

[0013] An artificial intelligence-based vehicle overall size rapid screening method, comprising the following steps:

[0014] Step S1, vehicle image and relative position information acquisition: through the high-definition camera and the laser ranging unit in the image acquisition module, the appearance image of the measured vehicle is collected from two angles of side view and top view, and the relative position information between the measured vehicle and the image acquisition module is synchronously acquired;

[0015] Step S2, vehicle type recognition: the deep learning model in the artificial intelligence recognition module is used to analyze the collected vehicle image, recognize the vehicle type, and acquire the overall size allowable range of the corresponding vehicle type according to the preset standard;

[0016] Step S3, vehicle overall size calculation: through the size calculation module, the mapping relationship between the image coordinate system and the world coordinate system is established by combining the vehicle image collected by the high-definition camera and the distance data measured by the laser ranging unit, the proportion of the image pixel size and the actual physical size is determined, and the actual overall size of the vehicle is calculated by using the image processing algorithm;

[0017] Step S4, Hui-shaped limit value space production: based on the vehicle overall size allowable range, the corresponding two-dimensional Hui-shaped space model is generated by using the space generation module, wherein the inner interface and the outer interface of the two-dimensional Hui-shaped space model correspond to the lower limit value and the upper limit value of the vehicle overall size respectively.

[0018] Step S5, eligibility screening judgment: through the screening judgment module, the calculated vehicle actual outer dimension is compared with the generated Hui-shaped space model, whether the vehicle actual outer dimension is completely located between the inner interface and the outer interface of the Hui-shaped space is judged, if the vehicle actual outer dimension meets the condition of not exceeding the inner interface and not exceeding the outer interface at the same time, it is determined that the screening is passed, if the vehicle actual outer dimension exceeds any one of the inner interface or the outer interface, it is determined that the screening is not passed;

[0019] Step S6, data storage and feedback: the screening result and data are stored in the database, which is convenient for subsequent query and analysis.

[0020] Further, the calculation process of the actual outer dimension of the measured vehicle in step S3 is specifically: based on the size of the sensor device of the high-definition camera in the two-dimensional plane of the imaging unit is L p ×D p , the image resolution size is x L ×x D , the camera focal length is f, the straight line distance measured by the laser ranging unit is h, the pixel size of the detected vehicle in the image is n L ×n D , the actual size of the detected vehicle is L×D×H.

[0021] In the top view, the actual length L of the detected vehicle is:

[0022]

[0023] In the top view, the actual width D of the detected vehicle is:

[0024]

[0025] In the side view, the actual height H of the detected vehicle is:

[0026]

[0027] In summary, due to the adoption of the above technical scheme, the beneficial effects of the present application are:

[0028] 1. In this invention, artificial intelligence is used to automatically identify vehicle types and calculate vehicle outline dimensions, greatly reducing the workload and time of manual measurement and improving detection efficiency. Furthermore, by utilizing deep learning models and image processing algorithms, vehicle types and vehicle outline dimensions can be identified more accurately, reducing measurement errors caused by human factors. At the same time, vehicle outline dimensions can be screened according to preset standards to ensure the legality and validity of the detection results. The entire system can complete the screening of vehicle outline dimensions in a short time and is suitable for scenarios requiring rapid detection, such as automobile production lines and vehicle inspection stations. It effectively solves the problems of low accuracy and low efficiency in existing vehicle outline dimension detection. Attached Figure Description

[0029] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort, wherein:

[0030] Figure 1 This is a schematic diagram of the system structure of the present invention;

[0031] Figure 2 This is a flowchart of the method of the present invention;

[0032] Figure 3 This is a schematic diagram of the screening and judgment module of the present invention screening the outer contour dimensions based on the U-shaped spatial model;

[0033] Figure 4 This is a schematic diagram illustrating how the size calculation module of this invention calculates the external dimensions of a car. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0035] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0036] It should be noted that the reference numerals and the letters in the following drawings represent similar items, and therefore, once an item is defined in one drawing, it need not be further defined and explained in subsequent drawings.

[0037] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship when the product of the application is usually placed, and are only for the convenience of the simplified description of the application, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the application. In addition, the terms "first", "second", "third" and the like are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0038] In addition, the terms "horizontal", "vertical" and the like do not mean that the components must be absolutely horizontal or vertical, but can be slightly inclined. For example, "horizontal" only means that its direction is more horizontal relative to "vertical", and does not mean that the structure must be completely horizontal, but can be slightly inclined.

[0039] In the description of the present application, it should also be noted that unless otherwise explicitly specified and limited, the terms "arrangement", "installation", "connection", "connection" should be broadly understood, for example, it can be fixedly connected, or it can be detachably connected or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the communication between two elements inside. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0040] The accompanying drawings incorporated in the specification Figure 1 ,

[0041] An artificial intelligence-based automobile external dimension rapid screening system, comprising an image acquisition module 1, an artificial intelligence recognition module 2, a dimension calculation module 3, a space generation module 4 and a screening judgment module 5.

[0042] The image acquisition module 1 is composed of a high-definition wide-angle camera and a laser ranging unit, adopts an integrated design, and is deployed on one side and the top of the vehicle detection channel. The high-definition wide-angle camera can synchronously acquire the appearance images of the vehicle from two angles of top view and side view, ensuring complete capture of the outer contour information of the vehicle being detected. The laser ranging unit is in the same position as the lens component in the high-definition wide-angle camera, and the laser ranging unit accurately projects the emitted laser beam to the surface of the vehicle shell and effectively reflects, thereby realizing high-precision distance measurement. The acquired vehicle appearance images and ranging data are sent in real time to the artificial intelligence recognition module 2 and the size calculation module 3 through a data transmission line or a wireless network, providing data support for subsequent vehicle type recognition and outer contour size calculation.

[0043] The artificial intelligence recognition module 2 adopts a deep learning model, including but not limited to target detection algorithm models such as YOLO and Faster R-CNN, to recognize the acquired vehicle images. The model is trained using a large amount of image data of different vehicle types in the training stage, and can accurately recognize various common vehicle types, including but not limited to common vehicle types such as flatbed trucks and van trucks. The recognition result includes the vehicle type and the outer contour size allowable range (mainly including the upper limit value and the lower limit value) of the corresponding vehicle type, and these upper and lower limit values are set according to relevant regulations such as the national standard GB1589-2016. The recognition result is sent in real time to the space generation module 4 through a data transmission line or a wireless network, providing data support for subsequent vehicle outer contour size compliance determination.

[0044] The size calculation module 3 receives the vehicle images and ranging data transmitted by the image acquisition module 1. The module first pre-processes the acquired images, including denoising, correction, and enhancement optimization operations to improve image quality. Subsequently, image processing algorithms, including edge detection, feature point extraction, and other operations, are used to analyze the vehicle images, combined with the distance data provided by the laser ranging unit, to establish the mapping relationship between the image coordinate system and the world coordinate system, determine the ratio of image pixel size to actual physical size, and accurately calculate the length, width, and height of the vehicle. After the calculation is completed, the actual outer contour size data of the vehicle is sent to the screening judgment module 5 through a data transmission line or a wireless network, providing accurate measurement results for subsequent vehicle outer contour size compliance determination.

[0045] The space generation module 4 generates a corresponding two-dimensional H-shaped space model according to the upper and lower limit values of the vehicle contour size provided by the artificial intelligence recognition module 2. The inner interface of the H-shaped space corresponds to the lower limit value of the vehicle contour size, and the outer interface corresponds to the upper limit value of the vehicle contour size. The H-shaped space model serves as a reference for vehicle contour size compliance screening, and quickly determines whether the actual vehicle contour size meets the requirements of the national standard GB1589-2016 through intuitive space boundary definition. The generation result is sent to the screening judgment module 5 through a data transmission line or a wireless network, providing clear judgment basis for the screening judgment module 5.

[0046] The screening judgment module 5 compares and analyzes the actual vehicle contour size provided by the size calculation module 3 with the H-shaped space provided by the space generation module 4. According to the relevant provisions in GB1589-2016, the size of the rearview mirror and other detachable parts of the vehicle type recognized by the artificial intelligence recognition module 2 is removed, and based on the processed vehicle size data, it is judged whether it is completely located between the inner interface and the outer interface of the H-shaped space, as shown in Figure 3 If the vehicle size does not exceed the inner interface and the outer interface of the H-shaped space, the screening is passed; otherwise, the screening is not passed. The screening result can be displayed in real time through the display screen, or transmitted to the related system through the network, for subsequent processing and data recording.

[0047] As shown in Figure 2 A vehicle contour size rapid screening method based on artificial intelligence, comprising the following steps:

[0048] Step S1, vehicle image and relative position information acquisition: through the high-definition camera and laser ranging unit in the image acquisition module, the appearance image of the measured vehicle is collected from two angles of side view and top view, and the relative position information between the measured vehicle and the image acquisition module is acquired synchronously;

[0049] Step S2, vehicle type recognition: using the pre-trained deep learning model in the artificial intelligence recognition module, the collected vehicle image is analyzed to recognize the vehicle type, and the contour size range of the corresponding vehicle type is obtained according to the national standard GB1589-2016;

[0050] Step S3, vehicle contour size calculation: through the size calculation module, combining the vehicle image collected by the high-definition camera and the distance data measured by the laser ranging unit, the mapping relationship between the image coordinate system and the world coordinate system is established, the proportion of the image pixel size and the actual physical size is determined, and the image processing algorithm is used to calculate the actual contour size of the vehicle;

[0051] Step S4, Production of the Rectangular Limit Space: Based on the allowable range of vehicle outline dimensions, a corresponding two-dimensional rectangular space model is generated using the space generation module. The inner and outer interfaces of the two-dimensional rectangular space model correspond to the lower and upper limits of the vehicle outline dimensions, respectively.

[0052] Step S5, Qualification Screening: The screening module compares the calculated actual vehicle dimensions with the generated U-shaped space model. According to the relevant provisions of GB1589-2016, after dimensional removal of the rearview mirrors and other detachable components, it determines whether the actual vehicle dimensions are completely within the inner and outer interfaces of the U-shaped space. If the actual vehicle dimensions simultaneously meet the condition of not exceeding the inner interface and not exceeding the outer interface, the screening is deemed successful; if the actual vehicle dimensions exceed either the inner or outer interface, the screening is deemed unsuccessful.

[0053] Step S6, Data storage and feedback: The screening results and data are stored in the database for subsequent querying and analysis. At the same time, the system can be optimized and adjusted based on the screening results to improve the accuracy and efficiency of screening.

[0054] Specifically, the calculation process of the actual outer dimensions of the vehicle under test in step S3 is as follows: the dimensions of the sensor device based on the high-definition camera in the two-dimensional plane of the imaging unit are L. p ×D p The image resolution size is x L ×x D The camera focal length is f, the straight-line distance between the laser ranging unit and the inspected vehicle is h, and the pixel size of the inspected vehicle in the image is n. L ×n D The actual dimensions of the inspected vehicle are L×D×H;

[0055] In the top view, the actual length L of the inspected vehicle is:

[0056]

[0057] In the top view, the actual width D of the inspected vehicle is:

[0058]

[0059] In the side view, the actual height H of the inspected vehicle is:

[0060]

[0061] like Figure 4As shown, the vehicle 11 in the world coordinate system is imaged by the light 12 through the lens component 13 of the high-definition camera, and a two-dimensional image is generated in the imaging unit of the high-definition camera, i.e., the vehicle 14 in the image coordinate system, the focal length of the camera is f, and the straight-line distance of the laser ranging unit to the vehicle under inspection is h. For the high-definition camera, the size of the sensor device in the two-dimensional plane of the imaging unit is L p ×D p , the image resolution is x L ×x D , the pixel size of the vehicle under inspection in the image is n L ×n D , and the actual size of the vehicle under inspection is L×D×H.

[0062] The above is the embodiment of the present application. The foregoing is the various preferred embodiments of the present application, and the preferred embodiments in the various preferred embodiments can be arbitrarily combined and used if not obviously contradictory or with a certain preferred embodiment as a prerequisite. The embodiments and the specific parameters in the embodiments are only for clearly describing the verification process of the application and are not intended to limit the patent protection scope of the present application. The patent protection scope of the present application is still subject to the claims, and any equivalent structural changes using the content of the specification and the drawings of the present application should also be included in the protection scope of the present application.

Claims

1. An artificial intelligence-based rapid screening system for the outer dimensions of a vehicle, characterized by, The application comprises an image acquisition module, an artificial intelligence recognition module, a size calculation module, a space generation module, and a screening judgment module. The image acquisition module is used to acquire the appearance image of the measured vehicle and the relative position information between the measured vehicle and the image acquisition module. The image acquisition module comprises a high-definition camera and a laser ranging unit. The high-definition camera is used to take the appearance image of the vehicle from two angles of side view and overhead view. The laser ranging unit is used to measure the straight-line distance between the measured vehicle and the image acquisition module. The artificial intelligence recognition module is in communication connection with the image acquisition module. The artificial intelligence recognition module is used to receive the vehicle image and the relative position information acquired by the image acquisition module, identify the vehicle type through a deep learning model according to the acquired vehicle image and the relative position information, and acquire the allowable range of the vehicle type according to a preset standard. The size calculation module is in communication connection with the image acquisition module. The size calculation module is used to receive the vehicle image and the relative position information acquired by the image acquisition module, and calculate the actual vehicle size according to the acquired vehicle image and the relative position information. The space generation module is in communication connection with the artificial intelligence recognition module. The space generation module is used to generate a two-dimensional back-shaped space model according to the allowable range of the vehicle size. The inner interface and the outer interface of the two-dimensional back-shaped space model correspond to the lower limit value and the upper limit value of the vehicle size, respectively. The screening judgment module is in communication connection with the size calculation module and the space generation module. The screening judgment module is used to compare and analyze the actual vehicle size output by the size calculation module with the two-dimensional back-shaped space model generated by the space generation module, and judge whether the actual vehicle size is within the allowable range of the vehicle size.

2. The artificial intelligence based rapid screening system for the outer dimensions of a vehicle as claimed in claim 1 wherein, The high-definition camera and the laser ranging unit of the image acquisition module are designed in an integrated manner. The laser ranging unit and the lens component in the high-definition camera are located at the same position.

3. An artificial intelligence-based rapid screening method for the external dimensions of a vehicle, characterized by, The application comprises the following steps: Step S1: vehicle image and relative position information acquisition. The high-definition camera and the laser ranging unit in the image acquisition module are used to acquire the appearance image of the measured vehicle from two angles of side view and overhead view, and to acquire the relative position information between the measured vehicle and the image acquisition module. Step S2: vehicle type identification. The deep learning model in the artificial intelligence recognition module is used to analyze the acquired vehicle image, identify the vehicle type, and acquire the allowable range of the vehicle size according to a preset standard. Step S3: vehicle size calculation. The mapping relationship between the image coordinate system and the world coordinate system is established by combining the vehicle image acquired by the high-definition camera and the distance data measured by the laser ranging unit. The proportion of the image pixel size and the actual physical size is determined. The image processing algorithm is used to calculate the actual vehicle size. Step S4: back-shaped limit value space production. The corresponding two-dimensional back-shaped space model is generated by the space generation module based on the allowable range of the vehicle size. The inner interface and the outer interface of the two-dimensional back-shaped space model correspond to the lower limit value and the upper limit value of the vehicle size, respectively. Step S5, eligibility screening judgment: through the screening judgment module, the calculated vehicle actual outer dimension is compared with the generated hui-shaped space model, whether the vehicle actual outer dimension is completely located between the inner interface and the outer interface of the hui-shaped space is judged, if the vehicle actual outer dimension meets the condition of not exceeding the inner interface and not exceeding the outer interface at the same time, it is determined that the screening passes; if the vehicle actual outer dimension exceeds any one of the inner interface or the outer interface, it is determined that the screening does not pass; Step S6, data storage and feedback: the screening result and data are stored in the database, which is convenient for subsequent query and analysis.

4. The method of claim 3, wherein the method is based on artificial intelligence. The calculation process of the actual outer dimension of the vehicle measured in the step S3 is as follows: the size of the sensor device of the high-definition camera in the length-width two-dimensional plane of the imaging unit is L p ×D p , the image resolution size is x L ×x D , the camera focal length is f, the straight-line distance measured by the laser ranging unit from the vehicle to be detected is h, the pixel size of the vehicle to be detected in the image is n L ×n D , and the actual size of the vehicle to be detected is L×D×H. In the top view, the actual length L of the vehicle under test is: In the top view, the actual width D of the vehicle under test is: In the side view, the actual height H of the vehicle under test is:

Images