Illegal parking detection linkage snapshot system and method

Abstract

The invention discloses an illegal parking detection linkage snapshot system and method, and relates to the technical field of image recognition. The method comprises the following steps: a roadside shooting device collects a video image of a current detection vision field; after the video image is preprocessed, feature point extraction is carried out on the image, and a white line parking space area on the road surface is obtained; identifying vehicles on the video image, and simulating three-dimensional positions of all vehicles in a detection vision field; and judging whether the three-dimensional simulation vehicle violates rules or not, generating a virtual three-dimensional image, and performing violation punishment. According to the invention, the two-dimensional images collected by the roadside shooting devices are converted into the three-dimensional point cloud data, the three-dimensional point cloud data generated by the shooting devices are spliced, and illegal parking judgment is carried out on each three-dimensional simulation vehicle, so that the illegal parking detection efficiency and the evidence obtaining and investigation accuracy are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of image recognition, and in particular relates to a system and method for detecting and capturing illegal parking. Background Art

[0002] With the development of social economy, the number of urban motor vehicles and motor vehicle drivers has increased rapidly. As a result, some motor vehicle drivers have a weak awareness of traffic safety, and the problem of urban parking difficulties has led to frequent illegal parking of motor vehicles on urban roads. Every year, the number of traffic congestion and car rear-end collisions caused by illegal parking of motor vehicles in the country is in the millions, which not only seriously affects the overall traffic environment of the city, but also causes personal and property losses to the people. At the same time, since the range of fixed cameras is very small, even if a parked vehicle is photographed, the camera cannot be controlled, the license plate number of the vehicle cannot be identified, and the violation penalty cannot be imposed, thus failing to play a role in supervision and illegal punishment. Therefore, timely and accurate evidence collection and investigation of illegal parking of motor vehicles is an important means to regulate drivers' driving behavior, ensure road traffic safety, and create a good urban traffic environment. Summary of the invention

[0003] The purpose of the present invention is to provide a system and method for detecting and capturing illegal parking, which solves the existing problems of difficult detection of illegal parking and inaccurate evidence collection and investigation by converting two-dimensional images collected by roadside shooting equipment into three-dimensional point cloud data, splicing the three-dimensional point cloud data generated by each shooting equipment, and making illegal parking judgments for each three-dimensional simulated vehicle.

[0004] In order to solve the above technical problems, the present invention is achieved through the following technical solutions: The present invention is a method for detecting and capturing illegal parking, comprising the following steps: Step S1: Add a camera to each road section, import the video stream into the illegal parking platform, set the capture angle of each camera, and start illegal parking detection; Step S2: setting illegal parking areas, parking areas and parking space areas on the ground of each road section; Step S3: The roadside shooting device collects the video image of the current detection field; Step S4: After preprocessing the video image, extract feature points from the image to obtain the white line parking space area on the road surface; Step S5: Identify the vehicles on the video image, simulate the contour moments of the vehicles, and simulate the three-dimensional positions of the contour moments of all vehicles in the detection field of view; Step S6: Determine whether the three-dimensional simulated vehicle violates traffic rules, and arrange law enforcement personnel to collect video; Step S7: The law enforcement officer uploads the collected video images to the illegal parking platform; Step S8: the illegal parking platform processes the video image and then splices it with the simulation platform; Step S9: Identify the license plate of the vehicle on the video image and match it with the vehicle simulated by the platform; Step S10: Generate a virtual three-dimensional image and impose penalties for violations.

[0005] As a preferred technical solution, in step S1, a camera is added to each road section to construct a spatial coordinate system; the spatial coordinate system includes world coordinates, camera coordinates, projection plane coordinates and pixel coordinates; the world coordinates are used to display the position of the vehicle in 3D space; the camera coordinates establish three-dimensional coordinates with the camera as the coordinate origin; the projection plane coordinates are used to describe the coordinate position of a 3D point in a spatial environment on the imaging plane of the camera after projection transformation; the pixel coordinates are used to represent a 2D image, each position of which represents a pixel point.

[0006] As a preferred technical solution, in step S2, illegal parking areas, parking areas and parking space areas are set on the ground of each road section, and the parking spaces on the ground are identified by an image recognition algorithm. The specific identification process is as follows: Step S21: converting the original image into an HSL image; Step S22: separating white from the HSL image; Step S23: combining the separated HSL image with the original image; Step S24: converting the combined image into a grayscale image; Step S25: using Gaussian blur algorithm to perform edge smoothing; Step S26: Apply the Canny edge detection algorithm on the smoothed gray image to determine the parking space area.

[0007] As a preferred technical solution, in step S4, the preprocessing process of the video image is as follows: Step S41: removing the noise generated when collecting the video image by using the neighborhood averaging method; the formula of the neighborhood averaging method is: ; In the formula, is the image after neighborhood averaging processing, is the number of row pixels, For images collected by the shooting device, The coordinates of neighboring pixels in the neighborhood; Step S42: using edge enhancement Sobel algorithm to perform edge enhancement processing; Step S43: Binarization is performed using a local optimal threshold method with a rotation in units of lines. The specific formula is as follows: ; In the formula, is the row pixel, is the maximum gray value in this row.

[0008] As a preferred technical solution, in step S5, all points on the vehicle contour are integrated to simulate the contour moment of the vehicle. Moment, the calculation formula is: ; In the formula, For the corresponding The moment in latitude, For the corresponding Moments in latitude; The contour moment performs binary image processing, treating all non-0 pixels as 1. The specific calculation formula is as follows: ; In the formula, for The average value of for The average value of .

[0009] As a preferred technical solution, in step S5, the coordinates of the multiple shooting devices are unified, the image pixels corresponding to the same vehicle are aligned with the color image pixels, the fusion of the multimodal feature image is completed, and the specific formula for transforming the pixel points to the world coordinates is: ; In the formula, is a constant coefficient, is the horizontal coordinate of the vehicle depth pixel coordinate, is the ordinate of the vehicle depth pixel coordinate, It is the actual size represented by the horizontal unit pixel on the photosensitive element of the shooting device. It is the actual size represented by the vertical unit pixel on the photosensitive element of the shooting device. is the pixel coordinate of the center of the image plane of the camera, is the focal length of the shooting device, is the rotation matrix between the camera coordinate system and the world coordinate system. is the translation matrix between the camera coordinate system and the world coordinate system. is the world coordinate corresponding to the pixel point; The specific formula for transforming two-dimensional pixels into three-dimensional point clouds is: ; In the formula, is the transformed point cloud coordinate, is the depth value, is the focal length of the shooting device, is the image coordinate corresponding to the pixel.

[0010] As a preferred technical solution, in step S6, before judging whether the three-dimensional simulated vehicle is in violation of traffic rules, it is first judged whether the vehicle is in a parked state; when the three-dimensional simulated vehicle of the identified vehicle does not move within three minutes, the vehicle is deemed to be parked; when the vehicle is in a parked state, it is judged whether the vehicle is in a parking area; when the vehicle is not in the parking area, it is directly judged that the vehicle is in a violation of traffic rules; when the vehicle is in the parking area, it is judged whether the vehicle is parked in the parking space area.

[0011] The present invention is a parking violation detection linkage capture system, comprising a parking violation platform, a roadside shooting device and a handheld device; The illegal parking platform includes a preprocessing module, a feature extraction module, a parking space recognition module, a three-dimensional model making module, a model splicing module and a violation penalty module; the preprocessing module is used to preprocess the video image collected by the shooting device; the feature extraction module is used to extract features from the preprocessed image; the parking space recognition module is used to identify the white line parking space area on the road surface; the three-dimensional model making module is used to convert the two-dimensional data of the video image into three-dimensional point cloud data; the model splicing module is used to model splice the three-dimensional point cloud data generated by each shooting device to generate a complete three-dimensional model of the road section; the violation penalty module is used to impose violation penalties on vehicles that park illegally; The video images captured by the roadside camera are uploaded to the illegal parking platform through the central processor; the data uploaded by the roadside camera also includes GPS positioning information, number, shooting height and shooting angle; The handheld device is used to photograph vehicle violation information and send the violation information to the illegal parking platform through a wireless transmission module.

[0012] The present invention has the following beneficial effects: The present invention converts two-dimensional images collected by roadside shooting equipment into three-dimensional point cloud data, splices the three-dimensional point cloud data generated by each shooting equipment, and makes illegal parking judgments on each three-dimensional simulated vehicle, thereby improving the efficiency of illegal parking detection and the accuracy of evidence collection and investigation.

[0013] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0015] Figure 1 A flow chart of a method for detecting and capturing illegal parking according to the present invention; Figure 2 It is a structural schematic diagram of a parking violation detection linkage capture system of the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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.

[0017] In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0018] In order to make the purpose, technical solution and advantages of this application clearer, the following Figure 2 The implementation methods of the present application are described in further detail.

[0019] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-2 It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application. Embodiment 1

[0020] See also Figure 1 As shown, the present invention is a method for detecting and capturing illegal parking, comprising the following steps: Step S1: Add a camera to each road section, import the video stream into the illegal parking platform, set the capture angle of each camera, and start illegal parking detection; Step S2: setting illegal parking areas, parking areas and parking space areas on the ground of each road section; Step S3: The roadside shooting device collects the video image of the current detection field; Step S4: After preprocessing the video image, extract feature points from the image to obtain the white line parking space area on the road surface; Step S5: Identify the vehicles on the video image, simulate the contour moments of the vehicles, and simulate the three-dimensional positions of the contour moments of all vehicles in the detection field of view; Step S6: Determine whether the three-dimensional simulated vehicle violates traffic rules, and arrange law enforcement personnel to collect video; Step S7: The law enforcement officer uploads the collected video images to the illegal parking platform; Step S8: the illegal parking platform processes the video image and then splices it with the simulation platform; Step S9: Identify the license plate of the vehicle on the video image and match it with the vehicle simulated by the platform; Step S10: Generate a virtual three-dimensional image and impose penalties for violations.

[0021] In step S1, a camera is added to each road section to construct a spatial coordinate system; the spatial coordinate system includes world coordinates, camera coordinates, projection plane coordinates and pixel coordinates; the world coordinates are used to display the position of the vehicle in 3D space; the camera coordinates establish three-dimensional coordinates with the camera as the coordinate origin; the projection plane coordinates are used to describe the coordinate position of a 3D point in the spatial environment on the imaging plane of the camera after projection transformation; the pixel coordinates are used to represent a 2D image, each position of which represents a pixel point.

[0022] In step S2, illegal parking areas, parking areas and parking space areas are set on the ground of each road section, and the parking spaces on the ground are identified by an image recognition algorithm. The specific identification process is as follows: Step S21: converting the original image into an HSL image; Step S22: separating white from the HSL image; Step S23: combining the separated HSL image with the original image; Step S24: converting the combined image into a grayscale image; Step S25: using Gaussian blur algorithm to perform edge smoothing; Step S26: Apply the Canny edge detection algorithm on the smoothed gray image to determine the parking space area.

[0023] In step S4, the preprocessing process of the video image is as follows: Step S41: removing the noise generated when collecting the video image by using the neighborhood averaging method; the formula of the neighborhood averaging method is: ; In the formula, is the image after neighborhood averaging processing, is the number of row pixels, For images collected by the shooting device, The coordinates of neighboring pixels in the neighborhood; Step S42: using edge enhancement Sobel algorithm to perform edge enhancement processing; Step S43: Binarization is performed using a local optimal threshold method with a rotation in units of lines. The specific formula is as follows: ; In the formula, is the row pixel, is the maximum gray value in this row.

[0024] Feature extraction uses edge detection algorithms such as Canny and Sobel to identify the edges of the car lines in the image. These algorithms can highlight the edges in the image and provide a basis for subsequent detection of whether the vehicle is in the parking space; multiple Sobel gradient filtering algorithms (such as x-axis direction, tangent direction, vector value size) can also be applied to filter out noise and enhance the features of the parking space line.

[0025] In step S5, all points on the vehicle contour are integrated to simulate the contour moment of the vehicle. Moment, the calculation formula is: ; In the formula, For the corresponding The moment in latitude, For the corresponding Moments in latitude; The contour moment performs binary image processing, treating all non-0 pixels as 1. The specific calculation formula is as follows: ; In the formula, for The average value of for The average value of .

[0026] The contour moment is calculated by integrating all points on the vehicle contour. It is a mathematical quantity used to describe the shape characteristics of the vehicle. The main functions of obtaining the contour moment of the vehicle are as follows: Different types of vehicles, such as cars, buses, trucks, etc., have different contour shapes. By calculating the contour moments of a vehicle, characteristic parameters that can characterize the shape of the vehicle can be extracted. Even for the same type of vehicle, different brands and models have unique appearance contours. Calculating contour moments can capture these subtle shape differences, thereby achieving more accurate brand and model recognition of the vehicle.

[0027] Calculating the contour moment of a vehicle can provide strong support for tracking the vehicle's trajectory. By continuously monitoring the changes in the contour moment of the vehicle and combining it with other image processing technologies, it is possible to accurately track the vehicle's motion trajectory and understand the vehicle's driving path, speed changes, and other information; The main function of this embodiment is to calculate the vehicle contour moment, which helps determine the accurate position and posture of the vehicle, such as the rotation angle and tilt degree of the vehicle. Taking the central moment as an example, it has a certain invariance to the transformations such as rotation, scaling and translation of the image. This feature can be used to calculate the angle information of the vehicle in the image, and then determine the driving direction and parking posture of the vehicle.

[0028] By determining the position information and parking posture of the vehicle in the image, we can roughly understand the proportion and position of the vehicle in the parking area in the image, so as to judge whether the vehicle is parked in a parking space, whether it is parked at an angle, whether one vehicle occupies multiple parking spaces, etc., which is convenient for judging whether the vehicle is parked illegally.

[0029] In step S5, the coordinates of the multiple shooting devices are unified, the image pixels corresponding to the same vehicle are aligned with the color image pixels, and the fusion of the multimodal feature image is completed. The specific formula for transforming the pixel points to the world coordinates is: ; In the formula, is a constant coefficient, is the horizontal coordinate of the vehicle depth pixel coordinate, is the ordinate of the vehicle depth pixel coordinate, It is the actual size represented by the horizontal unit pixel on the photosensitive element of the shooting device. It is the actual size represented by the vertical unit pixel on the photosensitive element of the shooting device. is the pixel coordinate of the center of the image plane of the camera, is the focal length of the shooting device, is the rotation matrix between the camera coordinate system and the world coordinate system. is the translation matrix between the camera coordinate system and the world coordinate system. is the world coordinate corresponding to the pixel point; The specific formula for transforming two-dimensional pixels into three-dimensional point clouds is: ; In the formula, is the transformed point cloud coordinate, is the depth value, is the focal length of the shooting device, is the image coordinate corresponding to the pixel.

[0030] The specific operation process is as follows: select a suitable world coordinate system as a reference, for example, take the first camera on the entire street as the origin, and determine the direction of the coordinate axis. Through the external parameter matrix of the camera, the coordinate system of each camera is unified into this world coordinate system, so that the coordinates of different cameras are comparable.

[0031] For images of the same vehicle taken with different cameras, the contour moments in the images are extracted. Then, the point pairs corresponding to the contour moments of the same vehicle in different images are found through contour moment matching. Based on the matched contour moment point pairs, the transformation matrix (such as affine transformation matrix or perspective transformation matrix) that transforms the pixels of one image to the pixel positions of another image is calculated using methods such as the RANSAC (random sampling consensus) algorithm. This transformation matrix can align the pixel positions of the same vehicle in different images.

[0032] Combined with the coordinate information of the camera obtained by calibration, the pinhole imaging model or a more complex imaging model is used to transform the pixels in the image into the world coordinate system. Specifically, for each pixel in the image, it is converted from pixel coordinates to camera coordinates, and then the camera coordinates are converted to world coordinates through the external parameter matrix.

[0033] In step S6, before determining whether the three-dimensional simulated vehicle is in violation of traffic rules, first determine whether the vehicle is in a parked state; when the three-dimensional simulated vehicle of the identified vehicle does not move within three minutes, the vehicle is deemed to be parked; when the vehicle is in a parked state, determine whether the vehicle is in the parking area; when the vehicle is not in the parking area, directly determine that the vehicle is in a violation of traffic rules; when the vehicle is in the parking area, determine whether the vehicle is parked in the parking space area.

[0034] In step S7, the law enforcement officer uploads the collected video image to the illegal parking platform. If the vehicle's license plate can be directly collected during the image processing, the violation penalty can be directly imposed. If the vehicle's license plate cannot be collected from the captured image, then after determining that the vehicle has violated the law, photos can be taken to collect evidence when the vehicle drives out of the parking space, and the law enforcement officer can be notified to go there to take photos to collect evidence.

[0035] In step S8, law enforcement personnel collect license plate information of the illegal vehicle, and splice it with the simulation platform after processing. The splicing method is consistent with the splicing method of the pictures collected by the above-mentioned multiple shooting devices, so it will not be repeated.

[0036] In step S9, after the license plate number is extracted, it is directly matched with the vehicle simulated by the platform, and the generated license plate is matched to the corresponding virtual vehicle, thereby generating a complete virtual three-dimensional simulated vehicle. Embodiment 2

[0037] See also Figure 2 As shown, the present invention is a parking violation detection linkage capture system, which can be used to execute the method content of embodiment 1 of the present invention, including a parking violation platform, a roadside shooting device and a handheld device; The illegal parking platform includes a preprocessing module, a feature extraction module, a parking space recognition module, a 3D model making module, a model splicing module and a violation penalty module; the preprocessing module is used to preprocess the video images collected by the shooting device; the feature extraction module is used to extract features from the preprocessed images; the parking space recognition module is used to identify the white line parking space area on the road surface; the 3D model making module is used to convert the 2D data of the video image into 3D point cloud data; the model splicing module is used to model splice the 3D point cloud data generated by each shooting device to generate a complete 3D model of the road section; the violation penalty module is used to impose violation penalties on vehicles that park illegally; The video images captured by the roadside camera are uploaded to the illegal parking platform through the central processor; the data uploaded by the roadside camera also includes GPS positioning information, number, shooting height and shooting angle; The handheld device is used to take pictures of vehicle violation information and send the violation information to the illegal parking platform through the wireless transmission module.

[0038] It is worth noting that in the above system embodiment, the various units included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0039] In addition, those skilled in the art can understand that all or part of the steps in the above-mentioned embodiments can be completed by instructing related hardware through a program, and the corresponding program can be stored in a computer-readable storage medium.

[0040] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for detecting and capturing illegal parking, characterized in that: The steps include: Step S1: Add a camera to each road section, import the video stream into the illegal parking platform, set the capture angle of each camera, and start illegal parking detection; Step S2: setting illegal parking areas, parking areas and parking space areas on the ground of each road section; Step S3: The roadside shooting device collects the video image of the current detection field; Step S4: After preprocessing the video image, extract feature points from the image to obtain the white line parking space area on the road surface; Step S5: Identify the vehicles on the video image, simulate the contour moments of the vehicles, and simulate the three-dimensional positions of the contour moments of all vehicles in the detection field of view; Step S6: Determine whether the three-dimensional simulated vehicle violates traffic rules, and arrange law enforcement personnel to collect video; Step S7: The law enforcement officer uploads the collected video images to the illegal parking platform; Step S8: the illegal parking platform processes the video image and then splices it with the simulation platform; Step S9: Identify the license plate of the vehicle on the video image and match it with the vehicle simulated by the platform; Step S10: Generate a virtual three-dimensional image and impose a penalty for traffic violation; In step S5, all points on the vehicle contour are integrated to simulate the contour moment of the vehicle. Moment, the calculation formula is: ; In the formula, For the corresponding The moment in latitude, For the corresponding Moments in latitude; The contour moment performs binary image processing, treating all non-0 pixels as 1. The specific calculation formula is as follows: ; In the formula, for The average value of for The average value of .

2. A method for detecting and capturing illegal parking according to claim 1, characterized in that: In the step S1, a camera is added to each road section to construct a spatial coordinate system; the spatial coordinate system includes world coordinates, camera coordinates, projection plane coordinates and pixel coordinates; the world coordinates are used to display the position of the vehicle in 3D space; the camera coordinates establish three-dimensional coordinates with the camera as the coordinate origin; the projection plane coordinates are used to describe the coordinate position of a 3D point in the spatial environment on the imaging plane of the camera after projection transformation; the pixel coordinates are used to represent a 2D image, each position of which represents a pixel point.

3. The method for detecting and capturing illegal parking according to claim 1, characterized in that: In step S2, illegal parking areas, parking areas and parking space areas are set on the ground of each road section, and the parking spaces on the ground are identified by an image recognition algorithm. The specific identification process is as follows: Step S21: converting the original image into an HSL image; Step S22: separating white from the HSL image; Step S23: combining the separated HSL image with the original image; Step S24: converting the combined image into a grayscale image; Step S25: using Gaussian blur algorithm to perform edge smoothing; Step S26: Apply the Canny edge detection algorithm on the smoothed gray image to determine the parking space area.

4. The method for detecting and capturing illegal parking according to claim 1, characterized in that: In step S4, the preprocessing process of the video image is as follows: Step S41: removing the noise generated when collecting the video image by using the neighborhood averaging method; the formula of the neighborhood averaging method is: ; In the formula, is the image after neighborhood averaging processing, is the number of row pixels, For images collected by the shooting device, are the coordinates of neighboring pixels in the neighborhood; Step S42: using edge enhancement Sobel algorithm to perform edge enhancement processing; Step S43: Binarization is performed using a local optimal threshold method with a rotation in units of lines. The specific formula is as follows: ; In the formula, is the row pixel, is the maximum gray value in this row.

5. The method for detecting and capturing illegal parking according to claim 1, characterized in that: In step S5, the coordinates of the multiple shooting devices are unified, the image pixels corresponding to the same vehicle are aligned with the color image pixels, and the fusion of the multimodal feature image is completed. The specific formula for transforming the pixel points to the world coordinates is: ； In the formula, is a constant coefficient, is the horizontal coordinate of the vehicle depth pixel coordinate, is the ordinate of the vehicle depth pixel coordinate, It is the actual size represented by the horizontal unit pixel on the photosensitive element of the shooting device. It is the actual size represented by the vertical unit pixel on the photosensitive element of the shooting device. is the pixel coordinate of the center of the image plane of the camera, is the focal length of the shooting device, is the rotation matrix between the camera coordinate system and the world coordinate system. is the translation matrix between the camera coordinate system and the world coordinate system. is the world coordinate corresponding to the pixel point; The specific formula for transforming two-dimensional pixels into three-dimensional point clouds is: ; In the formula, is the transformed point cloud coordinate, is the depth value, is the focal length of the shooting device, is the image coordinate corresponding to the pixel.

6. The method for detecting and capturing illegal parking according to claim 1, characterized in that: In step S6, before determining whether the three-dimensional simulated vehicle is in violation of traffic rules, it is first determined whether the vehicle is in a parked state; if the three-dimensional simulated vehicle of the identified vehicle does not move within three minutes, the vehicle is deemed to be parked; When the vehicle is in a parked state, determining whether the vehicle is in a parking area; When the vehicle is not in the parking area, it is directly determined that the vehicle is in a violation state; when the vehicle is in the parking area, it is determined whether the vehicle is parked in the parking space area.

7. A parking violation detection linkage capture system, comprising an illegal parking platform, a roadside shooting device and a handheld device, characterized in that: The illegal parking platform includes a preprocessing module, a feature extraction module, a parking space recognition module, a three-dimensional model making module, a model splicing module and a violation penalty module; the preprocessing module is used to preprocess the video images collected by the shooting device; the feature extraction module is used to extract features from the preprocessed images; The parking space recognition module is used to recognize the white line parking space area on the road surface; The three-dimensional model making module is used to convert the two-dimensional data of the video image into three-dimensional point cloud data; the model stitching module is used to stitch the three-dimensional point cloud data generated by each shooting device to generate a complete three-dimensional model of the road section; the violation penalty module is used to impose violation penalties on vehicles that park illegally; The video images captured by the roadside camera are uploaded to the illegal parking platform through the central processor; the data uploaded by the roadside camera also includes GPS positioning information, number, shooting height and shooting angle; The handheld device is used to photograph vehicle violation information and send the violation information to the illegal parking platform through a wireless transmission module.

Images