Transaction management system based on ETC control device

Abstract

The invention provides a transaction management system based on an ETC control device, and relates to the field of intelligent traffic, the transaction management system based on the ETC control device comprises a road side unit, a vehicle-mounted unit, a plurality of shooting units, a data processing unit and a center management unit, and the system carries out the image preprocessing of a plurality of collected images, screening the plurality of preprocessed images through a Brenner gradient function to obtain a first group of images, screening the plurality of preprocessed images through a Tengrad gradient function to obtain a second group of images, and performing intersection operation on the first group of images and the second group of images to obtain an initial processing image set; image registration and graph splicing are conducted on the initial processing image set, and a target image is obtained.When the vehicle is shot, the multiple shooting units are adopted to obtain multiple images, after the multiple images are obtained, the multiple images are screened and fused, clear images can be obtained, and accurate obtaining of vehicle information is facilitated.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation, and in particular to a transaction management system based on an ETC control device. Background Art

[0002] ETC, which means electronic toll collection, is an automatic toll collection system for highways or bridges. It uses a dedicated short-range communication between the on-board electronic tag installed on the windshield of the vehicle and the microwave antenna on the ETC lane at the toll station, and uses computer networking technology to conduct back-end settlement processing with the bank, so that vehicles can pay highway or bridge fees without stopping at the toll station.

[0003] When entering the toll station, the ETC will also take photos with the camera, record the license plate number, vehicle model and other information, accurately identify each vehicle passing through the ETC toll station, and ensure that the toll objects are accurate. However, in poor weather conditions, such as heavy rain, heavy snow, dense fog and other weather conditions, the transmission of microwave signals may be interfered with. These weather phenomena may cause microwave signal attenuation, scattering or reflection, thereby reducing the strength and stability of the signal, and further affecting the ETC microwave antenna's accurate recognition of the vehicle-mounted electronic tag. In addition, in poor weather conditions, the image information obtained by the camera unit may also be unclear, which, to a certain extent, affects the normal transaction of ETC vehicles.

[0004] Therefore, it is necessary to provide a new transaction management system based on ETC control device to solve the above technical problems. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides a transaction management system based on an ETC control device.

[0006] The transaction management system based on the ETC control device provided by the present invention comprises a roadside unit: installed on the ETC dedicated lane, responsible for sending a microwave signal to the vehicle-mounted unit to obtain the first vehicle information; On-board unit: installed in the vehicle using the ETC service, receives the microwave signal sent by the roadside unit, and transmits the first vehicle information back to the roadside unit through the microwave signal; Multiple camera units: installed on the ETC dedicated lanes to capture images of vehicles entering the predetermined area; Perform image preprocessing on the collected multiple images, screen the preprocessed multiple images through the Brenner gradient function to screen out a first group of images, screen the preprocessed multiple images through the Tenengrad gradient function to screen out a second group of images, perform intersection operation on the first group of images and the second group of images to obtain an initial processing image set, perform image registration and graphic splicing on the initial processing image set to obtain a target image; Data processing unit: used for receiving the first vehicle information acquired by the roadside unit and responsible for parsing the target image to acquire the second vehicle information; Central management unit: compares the first vehicle information and the second vehicle information, generates transaction information when the first vehicle information and the second vehicle information coincide, and sends the transaction information to the vehicle-mounted unit, otherwise, issues an alarm.

[0007] Preferably, the first vehicle information and the second vehicle information both include a license plate number, a vehicle type, and a vehicle model.

[0008] Preferably, the step of screening the preprocessed multiple images using the Brenner gradient function comprises: S301: convert multiple color images into grayscale images, grayscale value = 0.299R + 0.587G +0.114B; S302: traverse each pixel of the grayscale image to calculate the overall clarity of the image; D( )= ; in, Representing images Corresponding pixels Gray value, D( ) is the image clarity value; S303: Setting a threshold value for distinguishing a clear image from a blurred image; S304: Compare the clarity value of each image with a threshold value. Images with a value greater than the threshold value are clear images, which are screened out and marked as "satisfying clarity requirements". Images with a value less than the threshold value are blurred images.

[0009] Preferably, the step of screening the preprocessed multiple images using the Tenengrad gradient function comprises: S401: convert multiple color images into grayscale images, grayscale value = 0.299R + 0.587G +0.114B; S402: using the Sobel operator to calculate the gradient values ​​of each image in the horizontal and vertical directions respectively; S403: For each pixel point, calculate the square sum of its gradient values ​​in the horizontal and vertical directions to obtain the square sum of the gradient of the point; S404: traverse the entire image, accumulate the square sum of the gradients of all pixels, and then divide it by the total number of pixels to obtain an average gradient value of the entire image; S405: Setting a threshold value for distinguishing a clear image from a blurred image; S406: Compare the average gradient value of each image to be screened with the set threshold, screen out the clear images with average gradient values ​​greater than the threshold, and mark them as "satisfying the clarity requirement".

[0010] Preferably, the intersection operation comprises the following steps: S501: Create a new Boolean value array for storing images that meet the threshold requirements of both the Brenner gradient function and the Tenengrad gradient function; S502: for each image, simultaneously check the results of screening it using the Brenner gradient function and the Tenengrad gradient function; S503: If an image is marked as "satisfying the definition requirement" in the Brenner gradient function screening and is also marked as "satisfying the definition requirement" in the Tenengrad gradient function screening, then the image is added to the Boolean value array; otherwise, the image is ignored; S504: After traversing all images, the Boolean value array contains all images that meet the clarity requirements of the two functions at the same time, and the Boolean value array is output as the final screened clear image set, that is, the initial processed image set.

[0011] Preferably, the image registration and graphic stitching of the initially processed image set comprises the following steps: S601: Using SIFT algorithm, extracting feature points and their descriptors from the initial processed image set; When extracting feature points, a judgment threshold for the number of feature point extractions is set, and the contrast of the image is obtained through the gray level co-occurrence matrix. When the contrast is greater than the threshold, the number of extracted feature points is E, and when the contrast is less than the threshold, the number of extracted feature points is F, where E=2F.

[0012] S602: performing feature matching through feature points and their descriptors to find matching feature points between different images; S603: Calculate the geometric transformation relationship between the images in the initial processed image set according to the matched feature points, transform the images, and align the multiple images; S604: Use a weighted average algorithm to stitch the aligned images.

[0013] Preferably, the image preprocessing is noise reduction processing, and the noise reduction processing adopts any one of Gaussian filtering, mean filtering and non-local mean filtering.

[0014] Preferably, the image preprocessing is defogging processing, and the defogging processing includes the following steps: S801: grayscale processing of the image: converting the collected image into a grayscale image, and using a weighted average method to convert the RGB three channels of the color image into grayscale values; S802: extracting a dark channel: by calculating the minimum value of each pixel in the image in the three RGB color channels and finding the minimum value of the minimum value in the surrounding neighborhood of each pixel, a dark channel image is obtained; S803: Estimating atmospheric light: performing threshold processing on the dark channel image, calculating the color average of the corresponding points in the original color image of the pixels whose dark channel values ​​are higher than the threshold, as the estimated value of the atmospheric light; S804: Calculate the transmittance and obtain the defogging image.

[0015] Compared with the related art, the transaction management system based on the ETC control device provided by the present invention has the following beneficial effects: 1. The present invention screens out a first group of images through the Brenner gradient function, screens out a second group of images through the Tenengrad gradient function, and performs intersection operation to obtain a set of initially processed images. Since two algorithms are used to process images, the respective advantages of the Brenner gradient function and the Tenengrad gradient function are utilized, so that the obtained set of initially processed images is more stable.

[0016] 2. When photographing a vehicle, the present invention uses multiple photographing units to obtain multiple images, thereby increasing the reference basis for vehicle information. After obtaining multiple images, the multiple images are screened and fused, which is conducive to obtaining clearer images and facilitating the accurate acquisition of vehicle information. In addition, since screening is performed before the multiple images are fused, the speed of image fusion is improved to a certain extent.

[0017] 3. When managing transactions, the present invention reduces the error rate of transactions and possible subsequent problems by improving the clarity of images.

[0018] 4. When extracting feature points, the present invention sets a judgment threshold for the number of feature point extractions and obtains the contrast of the image through the grayscale co-occurrence matrix. When the contrast is greater than the threshold, the number of feature points extracted is greater than the number of feature points extracted when the contrast is less than the threshold. In this way, feature points can be extracted more precisely, thereby improving the matching accuracy to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A process wireframe diagram of a transaction management system based on an ETC control device provided by the present invention; Figure 2 A schematic diagram of the structure of the central management unit and the vehicle-mounted unit provided by the present invention; Figure 3 A flowchart of the steps of screening the Brenner gradient function provided by the present invention; Figure 4 A flowchart of the steps of screening the Tenengrad gradient function provided by the present invention; Figure 5 A flowchart of the intersection operation provided by the present invention; Figure 6 A wireframe diagram of the process of image registration and graphics stitching provided by the present invention; Figure 7 A wireframe diagram of the defogging process provided by the present invention. DETAILED DESCRIPTION

[0020] The present invention will be further described below in conjunction with the accompanying drawings and implementation modes.

[0021] Please refer to Figure 1-Figure 7 ,in, Figure 1 A process wireframe diagram of a transaction management system based on an ETC control device provided by the present invention; Figure 2 A schematic diagram of the structure of the central management unit and the vehicle-mounted unit provided by the present invention; Figure 3 A flowchart of the steps of screening the Brenner gradient function provided by the present invention; Figure 4 A flowchart of the steps of screening the Tenengrad gradient function provided by the present invention; Figure 5 A flowchart of the intersection operation provided by the present invention; Figure 6 A wireframe diagram of the process of image registration and graphics stitching provided by the present invention; Figure 7 A wireframe diagram of the defogging process provided by the present invention.

[0022] In the specific implementation process, Figure 1-Figure 7 As shown, Embodiment 1 The transaction management system based on the ETC control device includes: Roadside unit: installed on the ETC dedicated lane, responsible for sending microwave signals to the vehicle-mounted unit to obtain the first vehicle information; On-board unit: installed in the vehicle using the ETC service, receives the microwave signal sent by the roadside unit, and transmits the first vehicle information back to the roadside unit through the microwave signal; Multiple camera units: installed on the ETC dedicated lanes to capture images of vehicles entering the predetermined area; Perform image preprocessing on the collected multiple images, screen the preprocessed multiple images through the Brenner gradient function to screen out a first group of images, screen the preprocessed multiple images through the Tenengrad gradient function to screen out a second group of images, perform intersection operation on the first group of images and the second group of images to obtain an initial processing image set, perform image registration and graphic splicing on the initial processing image set to obtain a target image; Data processing unit: used for receiving the first vehicle information acquired by the roadside unit and responsible for parsing the target image to acquire the second vehicle information; Central management unit: compares the first vehicle information and the second vehicle information. When the first vehicle information and the second vehicle information coincide, generates transaction information and sends the transaction information to the vehicle-mounted unit. Otherwise, issues an alarm. Both the first vehicle information and the second vehicle information include the license plate number, vehicle type, and vehicle model. The steps of screening the preprocessed images by the Brenner gradient function include: S301: convert multiple color images into grayscale images, grayscale value = 0.299R + 0.587G +0.114B; S302: traverse each pixel of the grayscale image to calculate the overall clarity of the image; D( )= ; in, Representing images Corresponding pixels Gray value, D( ) is the image clarity value; S303: Setting a threshold value for distinguishing a clear image from a blurred image; S304: Compare the clarity value of each image with a threshold value, and select the image with a value greater than the threshold value as a clear image and mark it as "satisfying the clarity requirement", and the image with a value less than the threshold value as a blurred image; The steps of the Tenengrad gradient function to screen multiple preprocessed images include: S401: convert multiple color images into grayscale images, grayscale value = 0.299R + 0.587G +0.114B; S402: using the Sobel operator to calculate the gradient values ​​of each image in the horizontal and vertical directions respectively; S403: For each pixel point, calculate the square sum of its gradient values ​​in the horizontal and vertical directions to obtain the square sum of the gradient of the point; S404: traverse the entire image, accumulate the square sum of the gradients of all pixels, and then divide it by the total number of pixels to obtain an average gradient value of the entire image; S405: Setting a threshold value for distinguishing a clear image from a blurred image; S406: Compare the average gradient value of each image to be screened with the set threshold, screen out the clear images with average gradient values ​​greater than the threshold, and mark them as "satisfying the clarity requirement"; The intersection operation consists of the following steps: S501: Create a new Boolean value array for storing images that meet the threshold requirements of both the Brenner gradient function and the Tenengrad gradient function; S502: for each image, simultaneously check the results of screening it using the Brenner gradient function and the Tenengrad gradient function; S503: If an image is marked as "satisfying the definition requirement" in the Brenner gradient function screening and is also marked as "satisfying the definition requirement" in the Tenengrad gradient function screening, then the image is added to the Boolean value array; otherwise, the image is ignored; S504: After traversing all images, the Boolean value array contains all images that meet the clarity requirements of the two functions at the same time, and the Boolean value array is output as the final filtered clear image set, that is, the initial processed image set; The image registration and image stitching of the initially processed image set includes the following steps: S601: Using SIFT algorithm, extracting feature points and their descriptors from the initial processed image set; When extracting feature points, a judgment threshold for the number of feature point extractions is set, and the contrast of the image is obtained through the gray level co-occurrence matrix. When the contrast is greater than the threshold, the number of extracted feature points is E, and when the contrast is less than the threshold, the number of extracted feature points is F, where E=2F.

[0023] S602: performing feature matching through feature points and their descriptors to find matching feature points between different images; S603: Calculate the geometric transformation relationship between the images in the initial processed image set according to the matched feature points, transform the images, and align the multiple images; S604: Use a weighted average algorithm to stitch the aligned images. The image preprocessing is noise reduction processing. The noise reduction processing adopts any one of Gaussian filtering, mean filtering and non-local mean filtering.

[0024] Embodiment 2 Different from the first embodiment, the image preprocessing is defogging processing, and the defogging processing includes the following steps: S801: grayscale the image: convert the acquired image into a grayscale image, and the grayscale value is calculated as follows: Gray = 0.299R + 0.587G + 0.114B, where 0.299, 0.587 and 0.114 correspond to the weight coefficients of the red channel, the green channel and the blue channel respectively; S802: extracting a dark channel: by calculating the minimum value of each pixel in the image in the three RGB color channels and finding the minimum value of the minimum value in the surrounding neighborhood of each pixel, a dark channel image is obtained; S803: Estimating atmospheric light: performing threshold processing on the dark channel image, calculating the color average of the corresponding points in the original color image of the pixels whose dark channel values ​​are higher than the threshold, as the estimated value of the atmospheric light; S804: Calculate transmittance and obtain a defogging image; Establish atmospheric scattering model: I(x)=J(x)t(x)+A(1-(x)) Among them, I (x) is the collected image; J(x) is the dehazed image to be restored; t(x) is the transmittance; A is the estimated value of atmospheric light; Dark channel calculation method: (x) = ; Represents each channel of a color image, represents a local small area centered on pixel x, The gray value of the dark channel image is approximated to 0, then: →0; Normalize the model: =t(x) +1-t(X); Assuming that within a rectangular window Ω(x) of a certain size in the image, the value of t(x) is a constant t^(x), minimize both sides of the above equation, and then apply the dark channel prior to J, we get: (x) = =0; =0; Substituting the above formula into the minimization operation and introducing the parameter w=0.95, we can get t(x) as the transmittance: t(x)=1- ; J(x) = .

[0025] The circuits and controls involved in the present invention are all prior art and will not be described in detail here.

[0026] The above are only embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A transaction management system based on an ETC control device, characterized in that: include: Roadside unit: installed on the ETC dedicated lane, responsible for sending microwave signals to the vehicle-mounted unit to obtain the first vehicle information; On-board unit: installed in the vehicle using the ETC service, receives the microwave signal sent by the roadside unit, and transmits the first vehicle information back to the roadside unit through the microwave signal; Multiple camera units: installed on the ETC dedicated lanes to capture images of vehicles entering the predetermined area; Perform image preprocessing on the collected multiple images, screen the preprocessed multiple images through the Brenner gradient function to screen out a first group of images, screen the preprocessed multiple images through the Tenengrad gradient function to screen out a second group of images, perform intersection operation on the first group of images and the second group of images to obtain an initial processing image set, perform image registration and graphic splicing on the initial processing image set to obtain a target image; Data processing unit: used for receiving the first vehicle information acquired by the roadside unit and responsible for parsing the target image to acquire the second vehicle information; Central management unit: compares the first vehicle information and the second vehicle information, generates transaction information when the first vehicle information and the second vehicle information coincide, and sends the transaction information to the vehicle-mounted unit, otherwise, issues an alarm.

2. The transaction management system based on the ETC control device according to claim 1, characterized in that: The first vehicle information and the second vehicle information both include a license plate number, a vehicle type, and a vehicle model.

3. The transaction management system based on the ETC control device according to claim 1, characterized in that: The step of screening the preprocessed multiple images using the Brenner gradient function includes: S301: converting multiple color images into grayscale images; S302: traverse each pixel of the grayscale image to calculate the overall clarity of the image; S303: Setting a threshold value for distinguishing a clear image from a blurred image; S304: Compare the clarity value of each image with a threshold value. Images with a value greater than the threshold value are clear images, which are screened out and marked as "meeting the clarity requirement". Images with a value less than the threshold value are blurred images.

4. The transaction management system based on the ETC control device according to claim 3 is characterized in that: The step of screening the preprocessed multiple images by the Tenengrad gradient function includes: S401: converting multiple color images into grayscale images; S402: using the Sobel operator to calculate the gradient values ​​of each image in the horizontal and vertical directions respectively; S403: For each pixel point, calculate the square sum of its gradient values ​​in the horizontal and vertical directions to obtain the square sum of the gradient of the point; S404: traverse the entire image, accumulate the square sum of the gradients of all pixels, and then divide it by the total number of pixels to obtain an average gradient value of the entire image; S405: Setting a threshold value for distinguishing a clear image from a blurred image; S406: Compare the average gradient value of each image to be screened with the set threshold value, screen out the clear images with average gradient values ​​greater than the threshold value, and mark them as "satisfying the clarity requirement".

5. The transaction management system based on the ETC control device according to claim 4, characterized in that: The intersection operation comprises the following steps: S501: Create a new Boolean value array for storing images that meet the threshold requirements of both the Brenner gradient function and the Tenengrad gradient function; S502: for each image, simultaneously check the results of screening it using the Brenner gradient function and the Tenengrad gradient function; S503: If an image is marked as "satisfying the definition requirement" in the Brenner gradient function screening and is also marked as "satisfying the definition requirement" in the Tenengrad gradient function screening, then the image is added to the Boolean value array; otherwise, the image is ignored; S504: After traversing all images, the Boolean value array contains all images that meet the clarity requirements of the two functions at the same time, and the Boolean value array is output as the final screened clear image set, that is, the initial processed image set.

6. The transaction management system based on the ETC control device according to claim 1, characterized in that: The image registration and graphic stitching of the initially processed image set comprises the following steps: S601: Using SIFT algorithm, extracting feature points and their descriptors from the initial processed image set; When extracting feature points, a judgment threshold for the number of feature point extractions is set, and the contrast of the image is obtained through the gray level co-occurrence matrix. When the contrast is greater than the threshold, the number of extracted feature points is E, and when the contrast is less than the threshold, the number of extracted feature points is F, where E=2F.

7. S602: Perform feature matching through feature points and their descriptors to find matching feature points between different images; S603: Calculate the geometric transformation relationship between the images in the initial processed image set according to the matched feature points, transform the images, and align the multiple images; S604: Use a weighted average algorithm to stitch the aligned images.

8. The transaction management system based on the ETC control device according to claim 1, characterized in that: The image preprocessing is noise reduction processing, and the noise reduction processing adopts any one of Gaussian filtering, mean filtering and non-local mean filtering.

9. The transaction management system based on the ETC control device according to claim 1, characterized in that: The image preprocessing is defogging, and the defogging includes the following steps: S801: grayscale processing of the image: converting the acquired image into a grayscale image; S802: extracting a dark channel: by calculating the minimum value of each pixel in the image in the three RGB color channels and finding the minimum value of the minimum value in the surrounding neighborhood of each pixel, a dark channel image is obtained; S803: Estimating atmospheric light: performing threshold processing on the dark channel image, calculating the color average of the corresponding points in the original color image of the pixels whose dark channel values ​​are higher than the threshold, as the estimated value of the atmospheric light; S804: Calculate the transmittance and obtain the defogging image.

Images