An assisted driving safety system based on PMCW millimeter wave radar
By integrating PMCW millimeter wave radar and camera technology in the traffic system, the vehicle spacing and visibility data are corrected in real time, the problem of difficult vehicle spacing judgment under low visibility is solved, driving safety and traffic efficiency are improved, and the foundation for the development of intelligent transportation is laid. .
Patent Information
- Application Number
- CN202411216344.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-02
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-09-02
AI Technical Summary
Under abnormal weather conditions, such as heavy fog, heavy rain, and heavy snow, it is difficult for the driver to accurately judge the distance between vehicles, which increases the risk of traffic accidents.
Adopt an assisted driving safety system based on PMCW millimeter wave radar, through traffic recording module, traffic synchronization module, data correction module and vehicle early warning module, combined with high-precision millimeter wave radar and camera technology, real-time acquisition and correction of vehicle position and visibility data to achieve Accurate measurement and monitoring of vehicle spacing and issue early warnings if necessary.
It significantly improves the accuracy of vehicle spacing measurement and driving safety under low visibility conditions, reduces the risk of traffic accidents, and provides a foundation for the development of intelligent transportation systems.
Smart Images

Figure CN118736895B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of traffic control, and in particular to an auxiliary driving safety system based on PMCW millimeter wave radar. Background Art
[0002] In modern transportation systems, with the rapid increase in the number of motor vehicles, road safety has become a key issue of social concern. Traffic accidents occur frequently, especially in low-visibility environments such as fog, rain, and snow. The driver's vision is severely limited, and it becomes difficult to judge the distance between vehicles, which increases the risk of traffic accidents. Therefore, it is particularly important to develop an auxiliary system that can effectively improve road driving safety in low visibility conditions.
[0003] Millimeter-wave radar has gradually gained a place in vehicle assisted driving safety systems due to its excellent penetration and stable detection performance. In particular, millimeter-wave radar with PMCW (phase modulated continuous wave) technology is widely used in adaptive cruise control (ACC), collision warning, blind spot detection and other functions due to its high-precision distance and speed measurement capabilities.
[0004] In view of this, combining high-precision PMCW millimeter-wave radar technology with other sensors, introducing intelligent information processing and data fusion technology, and carrying out a series of innovations to make up for the shortcomings of the existing system have important practical significance for achieving the performance of vehicle assisted driving safety systems in abnormal weather conditions such as heavy fog, heavy rain, heavy snow and other low visibility conditions. Summary of the invention
[0005] The purpose of the present invention is to provide an assisted driving safety system based on PMCW millimeter wave radar to solve the following technical problems:
[0006] In abnormal weather conditions such as heavy fog, heavy rain, snow and other low visibility conditions, the vehicle assisted driving safety system is realized.
[0007] The purpose of the present invention can be achieved by the following technical solutions:
[0008] An assisted driving safety system based on PMCW millimeter wave radar, comprising:
[0009] Traffic recording module, which is used to collect the electronic license plate information of the vehicle through road facilities every time the vehicle enters the current road, record the vehicle model and license plate, obtain the vehicle's satellite positioning information and synchronize it to the cloud platform in real time;
[0010] Traffic synchronization module, used to obtain visibility information of the current road section through weather station data, obtain visibility distance, obtain two vehicles within the visibility distance error range through vehicle satellite positioning information, the error is preset, establish communication with the rear vehicle of the two vehicles, and apply for millimeter wave radar permission and camera permission of the rear vehicle;
[0011] The data correction module is used to measure the actual vehicle distance between the rear vehicle and the front vehicle through the millimeter wave radar of the rear vehicle, and correct the satellite positioning information according to the actual vehicle distance; capture the actual license plate image of the front vehicle through the camera of the rear vehicle, and extract the real-time license plate information according to the actual license plate image, compare the real-time license plate information with the recorded license plate information, judge the current actual visibility distance according to the image error of the license plate and the information error of the license plate information, and correct the current visibility information according to the actual visibility data;
[0012] The vehicle warning module is used to establish communication with the front and rear vehicles respectively when the safe vehicle distance corresponding to the actual visibility is greater than the current actual vehicle distance, and prompt the front and rear vehicles to increase the distance and maintain a safe speed.
[0013] As a further solution of the present invention: the data correction module also pre-processes the image taken by the camera, and the process is as follows:
[0014] The image captured by the camera is converted into a grayscale image through grayscale processing, and the grayscale image is subjected to median filtering. The grayscale image after median filtering is converted into a binary image, and the binary image is subjected to tilt correction and image segmentation.
[0015] As a further solution of the present invention: the process of the data correction module acquiring the actual license plate image is:
[0016] A rectangular frame is set, and the ratio of the long side to the short side of the rectangular frame is 44±a:14±a, where a is a preset error value. The image taken by the camera is traversed through the rectangular frame, and the black connected area in the binary image is framed to obtain several pending areas, and the ratio of the number of white pixels to the number of black pixels in each pending area is calculated. The pending area that meets the preset ratio threshold and the number of horizontal grayscale jumps is within [30,50] is selected, and the pending area is the actual license plate image obtained.
[0017] As a further solution of the present invention: the process of extracting real-time license plate information by the data correction module is:
[0018] Scan the actual license plate image horizontally, obtain the grayscale value of each column of pixels in turn, mark the white pixels as text pixels, and mark the black pixels as license plate background pixels. When the ratio of the number of text pixels to the total number of pixels in the current column is greater than the preset threshold, the column is the starting pixel column of the text. Continue to scan the actual license plate image until the column with the ratio of the number of text pixels to the total number of pixels in the current column is less than the preset threshold, which is the ending pixel column of the first character. Calculate the pixel distance between the starting pixel column and the ending pixel column. The pixel distance is the pixel width of a single character, marked as the character width. Continue to scan the actual license plate image, identify the characters whose ratio of the number of text pixels is greater than the preset threshold and whose pixel width is the character width, and segment each single character area to obtain several character areas.
[0019] As a further solution of the present invention: the process of the data correction module identifying image errors is:
[0020] A standard license plate image is generated based on the recorded license plate information. The actual license plate image and the standard license plate image are gray-scale processed and displayed in the same horizontal plane. Any pixel point in the actual license plate image is marked as a detection point. A first window W1 is established with the detection point as the center. The first window is in the positive direction and the pixel length of the side is an odd number. The grayscale values of all pixels in the first window W1 are obtained. The horizontal line where the detection point is located is obtained. The window at the same position in the standard license plate image is obtained along the horizontal line and marked as a second window W2. The grayscale values of all pixels in the second window W2 are obtained. The first window is overlapped with the second window, and the difference in the grayscale values of all overlapping pixels is calculated. The sum of the absolute values of the difference in the grayscale values is obtained, and two windows whose absolute value sum is greater than a preset threshold are marked as error areas. All windows of the actual license plate image and the standard license plate image are detected in turn, and the proportion of the error area to all windows is calculated, which is the image error.
[0021] As a further solution of the present invention: the process of calculating the information error by the data correction module is:
[0022] The real-time characters of the license plate are collected according to the real-time license plate information, the correct characters of the license plate are collected according to the recorded license plate information, and the ratio of the misrecognized character bytes to the total character bytes of the license plate is calculated, which is the information error.
[0023] As a further solution of the present invention: the process of determining the actual visibility distance is:
[0024] The information error is compared with the standard information error collected under the visibility of the meteorological station. When the information error is greater than the standard information error, the standard image error with a numerical difference of less than 10% from the image error is selected from the digital camera visibility system, and the visibility corresponding to the standard image error is taken as the actual visibility distance.
[0025] Beneficial effects of the present invention:
[0026] (1) The present invention integrates high-precision PMCW millimeter-wave radar and camera technology to achieve accurate measurement and monitoring of actual vehicle spacing. At the same time, the camera is used to obtain real-time license plate images, which are combined with satellite positioning information to accurately correct vehicle position and visibility data. This innovative solution significantly improves the accuracy and reliability of data, provides drivers with accurate driving assistance information, and effectively reduces the risk of potential traffic accidents under low visibility conditions;
[0027] (2) The system of the present invention uses an information synchronization module to obtain visibility information of the current road section, and combines millimeter-wave radar authority and camera authority to make real-time judgment and correction of the actual visibility distance. This unique data fusion method not only optimizes the driving experience and improves traffic efficiency, but also lays the foundation for the development of intelligent transportation systems. It has important practical significance and broad application prospects for improving road traffic safety, reducing energy consumption and environmental pollution, and promoting the development of intelligent transportation. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] The present invention will be further described below in conjunction with the accompanying drawings.
[0029] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION
[0030] 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.
[0031] See also Figure 1 As shown, the present invention is an assisted driving safety system based on PMCW millimeter wave radar, comprising:
[0032] Traffic recording module, which is used to collect the electronic license plate information of the vehicle through road facilities every time the vehicle enters the current road, record the vehicle model and license plate, obtain the vehicle's satellite positioning information and synchronize it to the cloud platform in real time;
[0033] Traffic synchronization module, used to obtain visibility information of the current road section through weather station data, obtain visibility distance, obtain two vehicles within the visibility distance error range through vehicle satellite positioning information, the error is preset, establish communication with the rear vehicle of the two vehicles, and apply for millimeter wave radar permission and camera permission of the rear vehicle;
[0034] The data correction module is used to measure the actual vehicle distance between the rear vehicle and the front vehicle through the millimeter wave radar of the rear vehicle, and correct the satellite positioning information according to the actual vehicle distance; capture the actual license plate image of the front vehicle through the camera of the rear vehicle, and extract the real-time license plate information according to the actual license plate image, compare the real-time license plate information with the recorded license plate information, judge the current actual visibility distance according to the image error of the license plate and the information error of the license plate information, and correct the current visibility information according to the actual visibility data;
[0035] The vehicle warning module is used to establish communication with the front and rear vehicles respectively when the safe vehicle distance corresponding to the actual visibility is greater than the current actual vehicle distance, and prompt the front and rear vehicles to increase the distance and maintain a safe speed.
[0036] By collecting and synchronizing vehicle information, visibility information, and satellite positioning data in real time, the relative position and distance between vehicles can be accurately determined. When the safe distance corresponding to the actual visibility is less than the current actual distance, the system will issue a warning in time to remind the driver to maintain a safe distance, thereby effectively reducing the occurrence of traffic accidents.
[0037] In another preferred embodiment of the present invention, the data correction module further preprocesses the image captured by the camera, and the process is as follows:
[0038] The image captured by the camera is converted into a grayscale image through grayscale processing, and the grayscale image is subjected to median filtering. The grayscale image after median filtering is converted into a binary image, and the binary image is subjected to tilt correction and image segmentation.
[0039] In another preferred situation of this embodiment, the process of the data correction module acquiring the actual license plate image is:
[0040] A rectangular frame is set, and the ratio of the long side to the short side of the rectangular frame is 44±a:14±a, where a is a preset error value. The image taken by the camera is traversed through the rectangular frame, and the black connected area in the binary image is framed to obtain several pending areas, and the ratio of the number of white pixels to the number of black pixels in each pending area is calculated. The pending area that meets the preset ratio threshold and the number of horizontal grayscale jumps is within [30,50] is selected, and the pending area is the actual license plate image obtained.
[0041] It is worth noting that the process of extracting real-time license plate information by the data correction module is:
[0042] Scan the actual license plate image horizontally, obtain the grayscale value of each column of pixels in turn, mark the white pixels as text pixels, and mark the black pixels as license plate background pixels. When the ratio of the number of text pixels to the total number of pixels in the current column is greater than the preset threshold, the column is the starting pixel column of the text. Continue to scan the actual license plate image until the column with the ratio of the number of text pixels to the total number of pixels in the current column is less than the preset threshold, which is the ending pixel column of the first character. Calculate the pixel distance between the starting pixel column and the ending pixel column. The pixel distance is the pixel width of a single character, marked as the character width. Continue to scan the actual license plate image, identify the characters whose ratio of the number of text pixels is greater than the preset threshold and whose pixel width is the character width, and segment each single character area to obtain several character areas.
[0043] In another preferred embodiment of the present invention, the process of the data correction module identifying image errors is:
[0044] A standard license plate image is generated based on the recorded license plate information. The actual license plate image and the standard license plate image are gray-scale processed and displayed in the same horizontal plane. Any pixel point in the actual license plate image is marked as a detection point. A first window W1 is established with the detection point as the center. The first window is in the positive direction and the pixel length of the side is an odd number. The grayscale values of all pixels in the first window W1 are obtained. The horizontal line where the detection point is located is obtained. The window at the same position in the standard license plate image is obtained along the horizontal line and marked as a second window W2. The grayscale values of all pixels in the second window W2 are obtained. The first window is overlapped with the second window, and the difference in the grayscale values of all overlapping pixels is calculated. The sum of the absolute values of the difference in the grayscale values is obtained, and two windows whose absolute value sum is greater than a preset threshold are marked as error areas. All windows of the actual license plate image and the standard license plate image are detected in turn, and the proportion of the error area to all windows is calculated, which is the image error.
[0045] In another preferred embodiment of the present invention, the process of calculating the information error by the data correction module is:
[0046] The real-time characters of the license plate are collected according to the real-time license plate information, the correct characters of the license plate are collected according to the recorded license plate information, and the ratio of the misrecognized character bytes to the total character bytes of the license plate is calculated, which is the information error.
[0047] In terms of judging the actual visibility distance, the present invention can effectively identify the actual visibility distance under the current environmental conditions by comparing the information error with the standard information error under the visibility of the meteorological station. This intelligent visibility evaluation mechanism enables the system to automatically adjust the safety warning parameters to ensure that accurate driving assistance information can be provided in different environments.
[0048] In another preferred embodiment of the present invention, the process of determining the actual visibility distance is:
[0049] The information error is compared with the standard information error collected under the visibility of the meteorological station. When the information error is greater than the standard information error, the standard image error with a numerical difference of less than 10% from the image error is selected from the digital camera visibility system, and the visibility corresponding to the standard image error is taken as the actual visibility distance.
[0050] In the data correction module, the information error between the real-time license plate information and the recorded license plate information is calculated to achieve a quantitative evaluation of the license plate recognition accuracy. This process improves the reliability of license plate recognition, ensures the accuracy of vehicle positioning and identity verification, and thus enhances the judgment and early warning capabilities of the entire system.
[0051] The present invention is an instrument system for automatically measuring visibility based on digital camera method. The digital camera directly captures the image of the selected target object and its background, and then transmits the image from the image acquisition card to the computer. By analyzing and processing the acquired image, the visibility value can be automatically obtained. The visibility value measured by this method is relatively close to the visibility value observed by laser radar.
[0052] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.
Claims
1. An assisted driving safety system based on PMCW millimeter wave radar, characterized in that: include: Traffic recording module, which is used to collect the electronic license plate information of the vehicle through road facilities every time the vehicle enters the current road, record the vehicle model and license plate, obtain the vehicle's satellite positioning information and synchronize it to the cloud platform in real time; Traffic synchronization module, used to obtain visibility information of the current road section through weather station data, obtain visibility distance, obtain two vehicles within the visibility distance error range through vehicle satellite positioning information, the error is preset, establish communication with the rear vehicle of the two vehicles, and apply for millimeter wave radar permission and camera permission of the rear vehicle; The data correction module is used to measure the actual vehicle distance between the rear vehicle and the front vehicle through the millimeter wave radar of the rear vehicle, and correct the satellite positioning information according to the actual vehicle distance; capture the actual license plate image of the front vehicle through the camera of the rear vehicle, and extract the real-time license plate information according to the actual license plate image, compare the real-time license plate information with the recorded license plate information, judge the current actual visibility distance according to the image error of the license plate and the information error of the license plate information, and correct the current visibility information according to the actual visibility data; The process of the data correction module identifying image errors is as follows: Generate a standard license plate image based on the recorded license plate information, grayscale the actual license plate image and the standard license plate image, display them in the same horizontal plane, mark any pixel in the actual license plate image as a detection point, establish a first window W1 with the detection point as the center, the first window is in the positive direction, and the pixel length of the side length is an odd number, obtain the grayscale values of all pixels in the first window W1, obtain the horizontal line where the detection point is located, obtain the window at the same position in the standard license plate image along the horizontal line and mark it as a second window W2, obtain the grayscale values of all pixels in the second window W2, overlap the first window with the second window, calculate the difference in grayscale values of all overlapping pixels, obtain the sum of the absolute values of the difference in grayscale values, mark two windows whose absolute value sum is greater than a preset threshold as error areas, detect all windows of the actual license plate image and the standard license plate image in turn, and calculate the proportion of the error area to all windows, which is the image error; The process of calculating the information error by the data correction module is as follows: Collect the real-time characters of the license plate according to the real-time license plate information, collect the correct characters of the license plate according to the recorded license plate information, and calculate the ratio of the wrongly recognized character bytes to the total character bytes of the license plate, which is the information error; The process of determining the actual visibility distance is: The information error is compared with the standard information error collected under the visibility of the meteorological station. When the information error is greater than the standard information error, a standard image error with a numerical difference of less than 10% with the image error is selected from the digital camera visibility system, and the visibility corresponding to the standard image error is taken as the actual visibility distance. The vehicle warning module is used to establish communication with the front and rear vehicles respectively when the safe vehicle distance corresponding to the actual visibility is greater than the current actual vehicle distance, and prompt the front and rear vehicles to increase the distance and maintain a safe speed.
2. The PMCW millimeter wave radar-based assisted driving safety system according to claim 1, characterized in that: The data correction module also pre-processes the image captured by the camera, and the process is as follows: The image captured by the camera is converted into a grayscale image through grayscale processing, and the grayscale image is subjected to median filtering. The grayscale image after median filtering is converted into a binary image, and the binary image is subjected to tilt correction and image segmentation.
3. The PMCW millimeter wave radar-based assisted driving safety system according to claim 2, characterized in that: The process of the data correction module acquiring the actual license plate image is as follows: A rectangular frame is set, and the ratio of the long side to the short side of the rectangular frame is 44±a:14±a, where a is a preset error value. The image taken by the camera is traversed through the rectangular frame, and the black connected area in the binary image is framed to obtain several pending areas, and the ratio of the number of white pixels to the number of black pixels in each pending area is calculated. The pending area that meets the preset ratio threshold and the number of horizontal grayscale jumps is within [30,50] is selected, and the pending area is the actual license plate image obtained.
4. The PMCW millimeter wave radar-based assisted driving safety system according to claim 3, characterized in that: The process of extracting real-time license plate information by the data correction module is as follows: Scan the actual license plate image horizontally, obtain the grayscale value of each column of pixels in turn, mark the white pixels as text pixels, and mark the black pixels as license plate background pixels. When the ratio of the number of text pixels to the total number of pixels in the current column is greater than the preset threshold, the column is the starting pixel column of the text. Continue to scan the actual license plate image until the column with the ratio of the number of text pixels to the total number of pixels in the current column is less than the preset threshold, which is the ending pixel column of the first character. Calculate the pixel distance between the starting pixel column and the ending pixel column. The pixel distance is the pixel width of a single character, marked as the character width. Continue to scan the actual license plate image, identify the characters whose ratio of the number of text pixels is greater than the preset threshold and whose pixel width is the character width, and segment each single character area to obtain several character areas.
