Vehicle bottom inspection imaging method and device of CMOS area-array camera

Abstract

The invention discloses a vehicle bottom inspection imaging method and device of a CMOS area-array camera, and relates to the technical field of image processing, and the method comprises the steps: a vehicle sensing device informs a vehicle sensing control module of a controller after sensing that a vehicle enters an inspection area; the vehicle sensing control module notifies the area-array camera to read data of the radar control module; the frame rate and the shutter speed of the CMOS for collecting the vehicle bottom image data are determined according to the speed data, the original image data collected by the area array vehicle bottom scanning camera are transmitted to the upper computer through the network, image correction, cutting and splicing are carried out, and a complete vehicle bottom image is formed and displayed. By reading the speed data of the vehicle, calculating the frame rate of the image data acquired by the camera CMOS matched with the current vehicle speed, and adjusting the frame rate of the camera CMOS and the shutter speed in real time, the quality of images imaged by the camera is higher, each image acquired by the CMOS is effective image data, the image post-processing is simpler and quicker, and the image processing time is shortened.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a vehicle underbody inspection imaging method and device using a CMOS area array camera. Background Art

[0002] With the development of the transportation industry and the continuous increase in the number of vehicles, public safety and vehicle maintenance have become increasingly important. In various vehicle inspection systems, under-vehicle inspection is a key link. Traditional under-vehicle inspection mainly relies on manual operation, which has the problems of low inspection efficiency, high cost and the inspection results being greatly affected by human factors. In recent years, with the advancement of image processing technology and computer vision technology, automatic under-vehicle inspection systems based on image processing have gradually become a hot spot for research and application. These systems use cameras to shoot under the vehicle, analyze and process the captured under-vehicle images through image processing technology, and can quickly and accurately detect abnormal conditions under the vehicle.

[0003] Existing undercar inspection imaging systems mainly use CCD (charge coupled device) cameras for imaging, but CCD cameras have disadvantages such as high cost, high power consumption and short working life. In contrast, CMOS (complementary metal oxide semiconductor) cameras have the advantages of low cost, low power consumption, fast response speed and long working life. Therefore, more and more undercar inspection systems have begun to use CMOS cameras for imaging. However, during the undercar inspection imaging process, due to the uncertainty of vehicle speed and the mismatch between the frame rate and shutter speed of CMOS, it is easy to cause poor imaging quality. In the prior art, the area array undercar scanning camera usually uses a fixed CMOS frame rate and shutter speed to perform undercar imaging, and cannot dynamically adjust the frame rate and shutter speed of the camera CMOS according to changes in vehicle speed, thereby affecting the imaging effect. Therefore, how to adjust the frame rate and shutter speed of the CMOS camera in real time according to the vehicle speed to improve the imaging quality has become a technical problem that needs to be solved urgently. Summary of the invention

[0004] Based on the above-mentioned shortcomings of the prior art, an object of the present invention is to provide a vehicle underbody inspection imaging method and device using a CMOS area array camera to solve the above-mentioned technical problems.

[0005] To achieve the above object, the present invention provides the following technical solutions: a vehicle underbody inspection imaging method using a CMOS array camera, which is applied to a vehicle underbody inspection imaging system using a CMOS array camera, the system comprising: a vehicle sensing device, a radar speed measuring device, a controller, an array camera, an LED intelligent fill light and a host computer, wherein the controller comprises a vehicle sensing control module, a radar control module and a fill light control module; the method comprises: S1: The vehicle sensing device senses that a vehicle has entered the inspection area and then notifies the vehicle sensing control module of the controller; S2: The fill light control module turns on the LED intelligent fill light, and the vehicle sensing control module notifies the area array camera to read the data from the radar control module; S3: The area array camera reads the speed data of the vehicle, and determines the frame rate and shutter speed of the area array camera CMOS image data acquisition according to the speed data; S4: The area array camera transmits the collected raw image data to the host computer through the network; S5: After the vehicle sensing device senses that the vehicle has left the inspection area, it notifies the vehicle sensing control module of the controller and notifies the fill light control module to turn off the LED intelligent fill light; S6: The vehicle sensing control module notifies the area array camera to stop taking pictures and stops reading data from the radar control module; S7: The host computer obtains all the raw image data from the area array camera, performs image correction, cropping and splicing, forms a complete vehicle bottom image and displays it.

[0006] The present invention is further configured to determine the frame rate and shutter speed of the area array camera CMOS image data acquisition according to the speed data, including: inputting the speed data into a trained decision tree model to obtain the frame rate and shutter speed of the CMOS, wherein the training logic of the decision tree model includes: Acquiring historical imaging data, wherein the imaging data includes speed data, frame rate, shutter speed, and image quality score; Preprocessing the historical imaging data, and dividing the preprocessed historical imaging data into a training set and a test set; A decision tree model is constructed using a CART algorithm, the speed data is used as an input feature, the frame rate and shutter speed are used as output features, and model parameters are set, wherein the model parameters include a maximum depth and a minimum number of sample splits; The decision tree model is trained using the training dataset, and the prediction performance of the decision tree model is verified using the test dataset.

[0007] The present invention is further configured that the use of a test data set to verify the prediction performance of the decision tree model comprises: Using the image quality score as the evaluation metric, calculate the comprehensive error score for predicting this frame rate and shutter speed; When the comprehensive error score of the decision tree model on the validation data set is less than or equal to the preset comprehensive error score threshold, it means that the decision tree model can stably output high-quality images at different vehicle speeds; when the comprehensive error score of the decision tree model on the validation data set is greater than the preset comprehensive error score threshold, the model parameters are reset.

[0008] The present invention is further configured such that the calculation logic of the comprehensive error score under the predicted frame rate and shutter speed is: ,in, is the comprehensive error score, is the root mean square error between the frame rate and shutter speed predicted by the model and the actual frame rate and shutter speed, is the average quality score of all images in the test dataset, is the maximum value of the image quality score. and is a weight parameter used to balance the effects of CMOS frame rate, shutter speed error, and image quality score.

[0009] The present invention is further configured such that the host computer performs image correction, cropping and splicing on all the original image data acquired from the area array camera to form a complete vehicle bottom image and display it, including: Correcting the original image data and constructing an image set according to an imaging sequence; Using an edge detection algorithm on the image data in the image set to determine a valid area of ​​the image data; Using a feature point detection algorithm to detect key feature points in a valid area of ​​image data adjacent to the image set; Matching key feature points of valid areas of adjacent image data of the image set using a feature matching algorithm; Perform image registration based on the matched key feature points and calculate the transformation matrix of adjacent image data; transforming all image data into the same coordinate system according to the transformation matrix; The edges of adjacent image data are fused and all image data are stitched into a complete bottom image of the vehicle.

[0010] The present invention is further configured such that the edge detection algorithm includes a Canny edge detection algorithm, the feature point detection algorithm includes a scale-invariant feature transformation algorithm, and the feature matching algorithm includes a fast nearest neighbor search algorithm.

[0011] The present invention is further configured that the method further comprises: Unify the color and brightness of the complete underbody image to ensure the consistency of the underbody image; Improve the quality of vehicle bottom images through image enhancement algorithms.

[0012] The present invention also provides a vehicle bottom inspection imaging device using a CMOS area array camera, the device comprising: First sensing module: the vehicle sensing device senses that a vehicle has entered the inspection area and then notifies the controller of the vehicle sensing control module; Speed ​​acquisition module: The fill light control module turns on the LED intelligent fill light, and the vehicle sensing control module notifies the area array camera to read the data from the radar control module; CMOS frame rate and shutter speed acquisition module: the area array camera 40 reads the speed data of the vehicle, and determines the frame rate and shutter speed of the area array camera 40 CMOS for collecting the image data of the bottom of the vehicle according to the speed data; Image upload module: The area array camera transmits the collected raw image data to the host computer through the network; Second sensing module: After the vehicle sensing device senses that the vehicle has left the inspection area, it notifies the vehicle sensing control module of the controller and notifies the fill light control module to turn off the LED intelligent fill light; Stop module: The vehicle sensing control module notifies the area array camera to stop taking pictures and stops reading data from the radar control module; Image processing module: The host computer obtains all the raw image data from the area array camera, performs image correction, cropping and splicing to form a complete underbody image and display it.

[0013] The present invention provides a vehicle bottom inspection imaging method and device using a CMOS area array camera. The method notifies a vehicle sensing control module of a controller after a vehicle sensing device senses that a vehicle enters an inspection area; a fill light control module turns on an LED intelligent fill light, and the vehicle sensing control module notifies the area array camera to read data from a radar control module; the area array camera reads the speed data of the vehicle, and determines the frame rate and shutter speed of the area array vehicle bottom scanning camera CMOS according to the speed data to continuously collect image data of the moving vehicle; the area array camera transmits the collected original image data to a host computer through a network; the vehicle sensing device notifies the vehicle sensing control module of the controller after sensing that the vehicle leaves the inspection area, and at the same time notifies the fill light control module to turn off the LED intelligent fill light; the vehicle sensing control module notifies the area array camera to stop taking pictures, and at the same time stops reading data from the radar control module; the host computer performs image correction, cropping and splicing on all original image data obtained from the area array camera to form a complete vehicle bottom image and display it, and the beneficial effects produced include: By using the trained decision tree model, the frame rate and shutter speed of CMOS are intelligently predicted according to the vehicle speed, ensuring that the vehicle can obtain the best image at different speeds. By introducing the comprehensive error score as an evaluation indicator, the prediction performance of the model is comprehensively evaluated through the combination of image quality score and the current CMOS frame rate and shutter speed error, ensuring the output of high-quality images. By collecting the vehicle speed and using it to control the frame rate and shutter speed of the CMOS sensor, each image taken by the camera is a valid image, making image processing simpler and faster, thereby reducing image processing time, shortening data transmission time, and ultimately speeding up the time to display a complete image of the vehicle chassis; The solution solves the problem that when the vehicle is slow or stopped, the CMOS camera of the vehicle under-scanning cameras on the market continues to shoot the same part of the vehicle under the vehicle with high frame rate and high shutter speed, thus generating a large amount of identical and invalid image data. At the same time, it solves the problem of occupying a large amount of system resources, alleviates the chip's cache space and processing memory, reduces the camera's power consumption, and extends the camera's service life.

[0014] By post-processing the image data, including image correction, edge detection, feature point detection, feature matching, image registration and stitching, the final vehicle bottom image is ensured to be complete and of high quality. By adopting the Canny edge detection algorithm, the scale-invariant feature transform (SIFT) algorithm and the fast nearest neighbor search (FLANN) algorithm, the accuracy and efficiency of image processing are improved.

[0015] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings: Figure 1 A schematic diagram of the structure of a vehicle underbody inspection imaging system using a CMOS area array camera according to an exemplary embodiment of the present invention; Figure 2 A flow chart of a vehicle underbody inspection imaging method using a CMOS area array camera is shown as an exemplary embodiment of the present invention; Figure 3 The present invention is a schematic structural diagram of a vehicle underbody inspection imaging device using a CMOS area array camera according to an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0017] The following will describe the embodiments of the present invention with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention, not for limiting the scope of protection of the present invention.

[0018] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0019] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it is obvious to those skilled in the art that the embodiments of the present invention can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present invention difficult to understand. Example 1

[0020] A vehicle underbody inspection imaging method using a CMOS area array camera, such as Figure 1 As shown, a vehicle underbody inspection imaging system applied to a CMOS area array camera comprises: a vehicle sensing device 10, a radar speed measuring device 20, a controller 30, an area array camera 40, an LED intelligent fill light 50 and a host computer 60, wherein the controller 30 comprises a vehicle sensing control module 31, a radar control module 32 and a fill light control module 33; Figure 2 As shown, the method includes: S1: The vehicle sensing device 10 senses that a vehicle has entered the inspection area and notifies the vehicle sensing control module 31 of the controller 30; S2: the fill light control module 33 turns on the LED intelligent fill light 50, and the vehicle sensing control module 31 notifies the area array camera 40 to read the data of the radar control module 32; S3: the area array camera 40 reads the speed data of the vehicle, and determines the frame rate and shutter speed of the CMOS image data collected by the area array camera 40 according to the speed data; S4: the area array camera 40 transmits the collected raw image data to the host computer 60 via the network; S5: After the vehicle sensing device 10 senses that the vehicle has left the inspection area, it notifies the vehicle sensing control module 31 of the controller 30 and notifies the fill light control module 33 to turn off the LED intelligent fill light 50; S6: The vehicle sensing control module 31 notifies the area array camera 40 to stop taking pictures and stops reading data from the radar control module 32; S7: The host computer 60 performs image correction, cropping and splicing on all the original image data acquired from the area array camera 40 to form a complete vehicle bottom image and display it.

[0021] Specifically, the physical implementation structure of this invention includes six main components: vehicle sensing device 10, radar speed measuring device 20, controller 30, array camera 40 (CMOS array undercarriage scanning camera), LED intelligent fill light 50, and host computer 60. The vehicle sensing device 10 is used to sense the entry and exit of the vehicle, the radar speed measuring device 20 is used to obtain the vehicle speed, the controller 30 includes a vehicle sensing control module 31, a radar control module 32 and a fill light control module 33, the vehicle sensing control module 31 is used to control and obtain vehicle sensing data, the radar control module 32 is used to control and obtain radar data, the fill light control module 33 is used to control the intelligent fill light module to perform intelligent fill light, the CMOS array undercarriage scanning camera is used to obtain image data, the LED intelligent fill light 50 is used for imaging fill light, and the host computer 60 is used to process the image data, and to control and display the undercarriage image for the user, and the above components are connected through a local area network.

[0022] Specifically, after the vehicle sensing device 10 senses that the vehicle has entered the inspection area, it notifies the vehicle sensing control module 31 of the controller 30, and the fill light control module 33 turns on the LED intelligent fill light 50. The vehicle sensing control module 31 notifies the CMOS area array vehicle bottom scanning camera to read the data of the radar control module 32. The CMOS area array vehicle bottom scanning camera reads the vehicle speed data, and inputs the vehicle speed into the trained decision tree model to obtain the frame rate and shutter speed of the CMOS, and sets it as the frame rate and shutter speed of the area array vehicle bottom scanning camera CMOS. The camera shoots and obtains image data according to the set CMOS parameters, and transmits it to the host computer 60 through the network. After the vehicle sensing device 10 senses that the vehicle has left the inspection area, it notifies the vehicle sensing control module 31 of the controller, and at the same time notifies the fill light control module 33 to turn off the LED intelligent fill light 50. The vehicle sensing control module 31 notifies the CMOS area array vehicle bottom scanning camera to stop taking pictures, and at the same time stops reading the vehicle speed of the radar. The host computer 60 post-processes the image data to form a complete vehicle bottom image and displays it.

[0023] The present invention is further configured to determine the frame rate and shutter speed of the CMOS image data collected by the area array camera 40 according to the speed data, including: inputting the speed data into a trained decision tree model to obtain the frame rate and shutter speed of the CMOS, wherein the training logic of the decision tree model includes: Obtain historical imaging data, wherein the imaging data includes speed data, frame rate, shutter speed and image quality score; specifically, set a set of standardized image quality score indicators, including but not limited to the following: clarity: measure the sharpness and detail resolution of the image; contrast: measure the contrast between the bright and dark parts of the image; noise level: measure the degree of random noise in the image; color accuracy: measure the accuracy of image color restoration; completeness: whether the image contains all the areas that should be photographed without being cut or omitted, and further, ensure that the scoring standards are consistent to avoid subjective factors affecting the scoring results. The specific scoring details are set according to specific needs and are not restricted here.

[0024] The historical imaging data is preprocessed, and the preprocessed historical imaging data is divided into a training set and a test set. In a feasible embodiment of the present invention, the division ratio is 70% for training and 30% for testing. It can also be set to 80% for training and 20% for testing, which is not limited here.

[0025] The decision tree model is constructed using the CART algorithm, the speed data is used as the input feature, the frame rate and shutter speed are used as the output feature, and the model parameters are set, wherein the model parameters include the maximum depth and the minimum number of sample splits; specifically, the CART (Classification and Regression Trees) algorithm is used to construct the decision tree model, the vehicle speed is used as the input feature, the frame rate and shutter speed of the CMOS are used as the output feature, and the model parameters are set: including the maximum depth (the maximum number of layers of the control tree) and the minimum number of sample splits (the minimum number of samples required for node splitting). The decision tree model predicts the output value by recursively splitting the data and finding the optimal splitting point. The maximum depth and minimum number of sample splits parameters control the complexity of the model to avoid overfitting or underfitting.

[0026] The decision tree model is trained using the training data set, and the prediction performance of the decision tree model is verified using the test data set. Specifically, the training set data is input into the decision tree model, and the model learns based on the relationship between vehicle speed and CMOS frame rate and shutter speed. The model finds the best splitting rule based on the frame rate, shutter speed and image quality score of CMOS at different speeds. Through the training data, the model can learn to select the appropriate frame rate and shutter speed at different speeds to improve the accuracy of the prediction.

[0027] The present invention is further configured that the use of a test data set to verify the prediction performance of the decision tree model comprises: The image quality score is used as an evaluation index to calculate the comprehensive error score under the predicted CMOS frame rate and shutter speed; the present invention is further configured such that the calculation logic of the comprehensive error score under the predicted frame rate and shutter speed is: ,in, is the comprehensive error score, is the root mean square error between the frame rate and shutter speed predicted by the model and the actual frame rate and shutter speed, is the average quality score of all images in the test dataset, is the maximum value of the image quality score. and is a weight parameter used to balance the effects of frame rate, shutter speed error and image quality score.

[0028] When the comprehensive error score of the decision tree model on the validation data set is less than or equal to the preset comprehensive error score threshold, it means that the decision tree model can stably output high-quality images at different vehicle speeds; when the comprehensive error score of the decision tree model on the validation data set is greater than the preset comprehensive error score threshold, the model parameters are reset. Specifically, a comprehensive error score threshold is preset to determine whether the model performance meets the requirements. When the comprehensive error score of the decision tree model on the validation data set is less than or equal to the preset comprehensive error score threshold, it means that the model can stably output high-quality images at different vehicle speeds. When the comprehensive error score is greater than the preset comprehensive error score threshold, it means that the model performance is poor and the model parameters need to be reset. The above logic ensures that the prediction performance of the model meets the expected requirements and can stably output high-quality images at different vehicle speeds.

[0029] The present invention is further configured such that the host computer 60 performs image correction, cropping and splicing on all the original image data acquired from the area array camera 40 to form a complete vehicle bottom image and display it, including: Correcting the original image data and constructing an image set in the order of shooting; specifically, performing geometric correction and illumination correction on the original image data to eliminate possible distortion and uneven illumination in the image, and organizing the corrected image data into an image set in the order of shooting for subsequent processing; Using an edge detection algorithm on the image data in the image set to determine the valid area of ​​the image data; specifically, using an edge detection algorithm (such as Canny edge detection) to process each image, extract edge information in the image, and determine the valid area of ​​the image, that is, the area containing important details and structures, based on the edge information; Using a feature point detection algorithm to detect key feature points in a valid area of ​​image data adjacent to the image set; specifically, using a feature point detection algorithm to detect key feature points in a valid area of ​​each image, the key feature points being unique points in the image, including corner points and edge intersections, to facilitate subsequent matching; A feature matching algorithm is used to match key feature points of valid areas of adjacent image data of the image set; specifically, the key feature points in adjacent images are matched using the feature matching algorithm, and the RANSAC algorithm is used during the matching process to eliminate false matches and improve the robustness of the matching.

[0030] Image registration is performed based on the matched key feature points, and a transformation matrix of adjacent image data is calculated; specifically, the transformation matrix is ​​used to align adjacent images into a unified coordinate system.

[0031] All image data are transformed into the same coordinate system according to the transformation matrix; the consistency of the images during the stitching process is ensured to avoid stitching errors, and a unified reference coordinate system is provided for image fusion and final stitching; The edges of adjacent image data are fused and all image data are spliced ​​into a complete vehicle bottom image. Specifically, the transformed image is edge-fused to eliminate the inconsistency of the seams and transition areas between adjacent images, and the image edges are processed using an image fusion algorithm.

[0032] The present invention is further configured such that the edge detection algorithm includes a Canny edge detection algorithm, the feature point detection algorithm includes a scale-invariant feature transformation algorithm, and the feature matching algorithm includes a fast nearest neighbor search algorithm. Specifically, the Canny edge detection algorithm is a multi-stage edge detection algorithm, which is intended to identify important edges in an image. The specific steps include: Gaussian filter denoising, image gradient calculation, non-maximum suppression, dual threshold detection and edge connection; providing high-precision edge detection, which can effectively distinguish between real edges and noise in an image, and the result of edge detection is used to determine the effective area of ​​the image, reducing the interference of irrelevant areas on subsequent processing; the scale-invariant feature transformation algorithm (SIFT) is used to detect the image. Key feature points can remain invariant under scale and rotation transformations. The specific steps include: constructing scale space, Gaussian difference (DoG) to detect key points, accurately locating key points, assigning directions, and feature descriptions; providing feature points with high repeatability and uniqueness, suitable for images of different scales and rotation angles; the fast nearest neighbor search algorithm (FLANN) is used to efficiently match feature points in images by quickly finding the nearest neighbor feature point pairs. The specific steps include: constructing an approximate nearest neighbor index, searching for nearest neighbor matching pairs, and applying the RANSAC algorithm to eliminate false matches; providing efficient and accurate feature point matching, capable of processing large-scale data, ensuring the reliability of matching, eliminating false matches, and improving the accuracy of image registration.

[0033] The present invention is further configured that the method further comprises: Unify the color and brightness of the complete underbody image to ensure the consistency of the underbody image; specifically, color correction uses a color balance algorithm, including white balance adjustment and histogram equalization, to perform color correction on the spliced ​​underbody image to ensure the uniformity and naturalness of the color in the image and avoid color differences between different images; brightness correction uses a brightness adjustment algorithm, including histogram equalization and gamma correction, to perform brightness correction on the spliced ​​image to ensure the consistency of the brightness of the entire image without too dark or too bright areas.

[0034] The quality of the image under the vehicle is improved by an image enhancement algorithm. The image enhancement algorithm includes sharpening processing, denoising processing and contrast enhancement. The sharpening algorithm is used to sharpen the image to improve the details and clarity of the image, making the details of the underside of the vehicle clearer; the denoising algorithm is used to denoise the image to reduce the noise in the image and improve the quality of the image; the contrast enhancement algorithm is used to enhance the contrast of the image to improve the contrast of the image and make the details more obvious. The above-mentioned technology is existing technology and is not limited here. Example 2

[0035] See also Figure 3 , the exemplary vehicle underbody inspection imaging device of a CMOS area array camera comprises: First sensing module: the vehicle sensing device 10 senses that a vehicle has entered the inspection area and then notifies the vehicle sensing control module 31 of the controller 30; Speed ​​acquisition module: the fill light control module 33 turns on the LED intelligent fill light 50, and the vehicle sensing control module 31 notifies the area array camera 40 to read the data of the radar control module 32; CMOS frame rate and shutter speed acquisition module: the area array camera 40 reads the speed data of the vehicle, and determines the frame rate and shutter speed of the CMOS image data collected by the area array camera 40 according to the speed data; Image upload module: the area array camera 40 transmits the captured raw image data to the host computer 60 through the network; Second sensing module: after the vehicle sensing device 10 senses that the vehicle has left the inspection area, it notifies the vehicle sensing control module 31 of the controller 30 and notifies the fill light control module 33 to turn off the LED intelligent fill light 50; Stop module: the vehicle sensing control module 31 notifies the area array camera 40 to stop taking pictures and stops reading data from the radar control module 32; Image processing module: The host computer 60 obtains all the original image data from the area array camera 40, performs image cropping and splicing, forms a complete vehicle bottom image and displays it.

[0036] It should be noted that the undercarriage inspection imaging device of a CMOS area array camera provided in the above embodiment and the undercarriage inspection imaging method of a CMOS area array camera provided in the above embodiment belong to the same concept, wherein the specific manner in which each module and unit performs the operation has been described in detail in the method embodiment and will not be repeated here. In actual application, the undercarriage inspection imaging device of a CMOS area array camera provided in the above embodiment can allocate the above functions to different functional modules as needed, that is, divide the internal structure of the system into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0037] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0038] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0039] In this application, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0040] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0041] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0042] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0043] In the several embodiments provided in the present application, it should be understood that the disclosed system can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0044] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0045] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0046] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0047] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A vehicle underbody inspection imaging method using a CMOS area array camera. It is characterized in that A vehicle underbody inspection imaging system applied to a CMOS area array camera, the system comprising: a vehicle sensing device (10), a radar speed measurement device (20), a controller (30), an area array camera (40), an LED intelligent fill light (50) and a host computer (60), wherein the controller (30) comprises a vehicle sensing control module (31), a radar control module (32) and a fill light control module (33); the method comprises: S1: the vehicle sensing device (10) senses that a vehicle has entered the inspection area and notifies the vehicle sensing control module (31) of the controller (30); S2: the fill light control module (33) turns on the LED intelligent fill light (50), and the vehicle sensing control module (31) notifies the area array camera (40) to read the data from the radar control module (32); S3: the area array camera (40) reads the speed data of the vehicle, and determines the frame rate and shutter speed of the CMOS for collecting image data by the area array camera (40) according to the speed data; S4: the area array camera (40) transmits the collected raw image data to the host computer (60) via the network; S5: After the vehicle sensing device (10) senses that the vehicle has left the inspection area, it notifies the vehicle sensing control module (31) of the controller (30), and at the same time notifies the fill light control module (33) to turn off the LED intelligent fill light (50); S6: the vehicle sensing control module (31) notifies the area array camera (40) to stop taking pictures and stops reading data from the radar control module (32); S7: The host computer (60) performs image correction, cropping and splicing on all the original image data acquired from the area array camera (40) to form a complete vehicle bottom image and display it.

2. The vehicle underbody inspection imaging method using a CMOS area array camera according to claim 1, It is characterized in that Determining the frame rate and shutter speed of the CMOS of the area array camera (40) according to the speed data comprises: inputting the speed data into a trained decision tree model to obtain the frame rate and shutter speed of the CMOS, wherein the training logic of the decision tree model comprises: Acquiring historical imaging data, wherein the imaging data includes speed data, frame rate, shutter speed, and image quality score; Preprocessing the historical imaging data, and dividing the preprocessed historical imaging data into a training set and a test set; A decision tree model is constructed using a CART algorithm, the speed data is used as an input feature, the frame rate and the shutter speed are used as output features, and model parameters are set, wherein the model parameters include a maximum depth and a minimum number of sample splits; The decision tree model is trained using the training data set, and the prediction performance of the decision tree model is verified using the test data set.

3. The vehicle underbody inspection imaging method using a CMOS area array camera according to claim 2, It is characterized in that The use of the test data set to verify the prediction performance of the decision tree model includes: Using the image quality score as the evaluation metric, the comprehensive error score under the predicted frame rate and shutter speed is calculated; When the comprehensive error score of the decision tree model on the validation data set is less than or equal to the preset comprehensive error score threshold, it means that the decision tree model can stably output high-quality images at different vehicle speeds; when the comprehensive error score of the decision tree model on the validation data set is greater than the preset comprehensive error score threshold, the model parameters are reset.

4. The vehicle underbody inspection imaging method using a CMOS area array camera according to claim 3, It is characterized in that The calculation logic of the comprehensive error score under the predicted frame rate and shutter speed is: ,in, is the comprehensive error score, is the root mean square error between the frame rate and shutter speed predicted by the model and the actual frame rate and shutter speed, is the average quality score of all images in the test dataset, is the maximum value of the image quality score. and is a weight parameter used to balance the impact of frame rate and shutter speed errors and image quality scores.

5. The vehicle underbody inspection imaging method using a CMOS area array camera according to claim 1, It is characterized in that The host computer (60) performs image correction, cropping and splicing on all the raw image data acquired from the area array camera (40) to form a complete vehicle bottom image and display it, including: Correcting the original image data and constructing an image set according to the shooting order; Using an edge detection algorithm on the image data in the image set to determine a valid area of ​​the image data; Using a feature point detection algorithm to detect key feature points in a valid area of ​​image data adjacent to the image set; Matching key feature points of valid areas of adjacent image data of the image set using a feature matching algorithm; Perform image registration based on the matched key feature points and calculate the transformation matrix of adjacent image data; transforming all image data into the same coordinate system according to the transformation matrix; The edges of adjacent image data are fused and all image data are stitched into a complete bottom image of the vehicle.

6. The vehicle underbody inspection imaging method using a CMOS area array camera according to claim 5, It is characterized in that The edge detection algorithm includes a Canny edge detection algorithm, the feature point detection algorithm includes a scale-invariant feature transformation algorithm, and the feature matching algorithm includes a fast nearest neighbor search algorithm.

7. The vehicle underbody inspection imaging method using a CMOS area array camera according to claim 1, It is characterized in that The method further comprises: Unify the color and brightness of the complete underbody image to ensure the consistency of the underbody image; Improve the quality of vehicle bottom images through image enhancement algorithms.

8. A CMOS area array camera for underbody inspection imaging. It is characterized in that include: First sensing module: a vehicle sensing control module (31) that notifies the controller (30) after the vehicle sensing device (10) senses that the vehicle has entered the inspection area; Speed ​​acquisition module: the fill light control module (33) turns on the LED intelligent fill light (50), and the vehicle sensing control module (31) notifies the area array camera (40) to read the data of the radar control module (32); CMOS frame rate and shutter speed acquisition module: the area array camera 40 reads the speed data of the vehicle, and determines the frame rate and shutter speed of the area array camera 40 CMOS for collecting image data according to the speed data; Image upload module: the area array camera (40) transmits the collected raw image data to the host computer (60) via the network; Second sensing module: after the vehicle sensing device (10) senses that the vehicle has left the inspection area, it notifies the vehicle sensing control module (31) of the controller (30), and at the same time notifies the fill light control module (33) to turn off the LED intelligent fill light (50); Stop module: The vehicle sensing control module (31) notifies the area array camera (40) to stop taking pictures and stops reading data from the radar control module (32); Image processing module: The host computer (60) performs image correction, cropping and splicing on all the original image data acquired from the area array camera (40) to form a complete vehicle bottom image and display it.

Images