Visual identification system for automobile auxiliary driving

By designing a visual recognition system in the car assisted driving system, real-time acquisition and image enhancement, the problem of image quality degradation in harsh climates is solved, and the driver's road condition recognition accuracy and driving safety are improved.

CN120191290AActive Publication Date: 2025-06-24SHENZHEN XIARUI INTELLIGENT VISION TECHNOLOGY CO LTD

Patent Information

Application Number
CN202510270407.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-03-07
Publication Date
2025-06-24
Estimated Expiration
2045-03-07

AI Technical Summary

Technical Problem

The existing automotive assisted driving system is difficult to adapt in real time in harsh climates and complex environments, resulting in image quality degradation and processing delays, affecting driving safety.

Method used

A visual recognition system for automobile assisted driving is designed to acquire road condition images in real time through an image acquisition device, and use a processor to perform image enhancement, including adaptive contrast limitation and texture confidence analysis, to generate enhanced road condition images to assist driving.

Benefits of technology

The contrast of images and the obviousness of road conditions are improved, the accuracy of road conditions recognition and driving safety of drivers are enhanced, and the adaptability and real-timeness are improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120191290A_ABST
    Figure CN120191290A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of automobile auxiliary driving, in particular to a visual identification system for automobile auxiliary driving, which comprises an image acquisition device arranged outside an automobile and used for acquiring road condition images shot by the automobile in real time; the processor is connected with the image acquisition device and the display device and is used for receiving the road condition image sent by the image acquisition device, performing image enhancement on the road condition image to obtain an enhanced road condition image and sending the enhanced road condition image to the display device; and the display device is arranged on the automobile instrument desk, faces the driver and is used for receiving and displaying the enhanced road condition image so as to assist the driver in driving and improve the road condition recognition accuracy and driving safety of the driver.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of automotive assisted driving, and particularly to a vision recognition system for automotive assisted driving. Background Art

[0002] With the development of automotive assisted driving technology, vision perception systems have become one of the key technologies for realizing intelligent driving. Currently, the mainstream radar and vision fusion systems adopt a multi-sensor fusion method, specifically through the combination of sensors including but not limited to high-resolution cameras, millimeter-wave radars, and lidar, etc., to achieve environmental perception and driving assistance functions, showing a certain degree of reliability. However, this system also has some defects:

[0003] 1. Its multi-sensor configuration results in high hardware costs, and at the same time requires complex hardware interfaces and system integration solutions. In addition, high-precision sensors such as lidar are expensive, and the repair and maintenance costs are also relatively high, making it difficult for the radar and vision fusion system to be widely applied to mid- and low-end vehicles;

[0004] 2. The radar and vision fusion system needs to process heterogeneous data from multiple sources simultaneously, including but not limited to the distance information of the radar and the image data of the camera. These heterogeneous data usually have different sampling frequencies, resolutions, and error characteristics, resulting in frequent information conflicts between multi-sensor data. For example, the object distance detected by the radar does not match the object position recognized by the camera, making it difficult for the system to correctly judge the road environment; secondly, when fusing multi-sensor data, the fusion algorithm depends on a large amount of computing resources, especially in a real-time environment, a high-performance processor is required to complete data integration, filtering, and decision-making, increasing the computing burden and response time of the system. Therefore, complex time synchronization and spatial correction are required when fusing multi-sensor data.

[0005] 3. Under harsh climate conditions such as heavy rain, snow, and sandstorms, the performance of both the radar and the camera will be significantly affected. For example, the radar signal may be interfered by raindrops or snowflakes, and the camera is prone to overexposure or underexposure of images in scenes with drastic light changes (such as the entrances and exits of tunnels), making the system prone to making wrong judgments due to abnormal data, thus increasing the driving risk.

[0006] In the prior art, in order to overcome the defects of the radar and vision fusion system, a pure vision assistance system is provided. This system only collects external images outside the vehicle through a camera during driving to achieve environmental perception and driving assistance functions. However, during driving, due to the influence of weather conditions, such as rain, snow, sand, wind, and sunlight, the quality of the acquired external images will decline, and processing high-resolution images requires a large amount of computing power, making it prone to processing delays in complex environments, reducing the accuracy of road condition analysis, being unable to provide effective driving assistance information in real time, and affecting driving safety. Summary of the Invention

[0007] In view of this, embodiments of the present invention provide a vision recognition system for automotive assisted driving to solve the problems of poor environmental adaptability and real-time performance of the pure vision assistance system in the prior art.

[0008] Embodiments of the present invention provide a vision recognition system for automotive assisted driving, which includes:

[0009] An image acquisition device, arranged outside the vehicle, for real-time acquisition of road condition images captured by the vehicle;

[0010] A processor, connected to the image acquisition device and the display device, for receiving the road condition images sent by the image acquisition device, performing image enhancement on the road condition images to obtain enhanced road condition images, and sending the enhanced road condition images to the display device;

[0011] A display device, arranged on the vehicle dashboard and facing the driver, for receiving and displaying the enhanced road condition images to assist the driver in driving.

[0012] Further, the performing image enhancement on the road condition images to obtain enhanced road condition images includes:

[0013] Dividing the road condition images into blocks to obtain different road condition regions;

[0014] Obtaining different gray ranges according to the change characteristics of the number of pixel points in the gray histogram of the road condition region; obtaining a gray distribution characteristic value according to the difference characteristics of the number of pixel points with different grays in the gray range; obtaining a contrast characteristic value according to the gray characteristics of the road condition region, the quantity characteristics of the gray range, and the gray distribution characteristic value;

[0015] Cluster according to the gradient features of the pixel points in the road condition area to obtain two pixel point clusters; obtain the texture confidence according to the gradient difference features of the pixel point clusters and the distance features of the pixel points in the pixel point clusters; obtain the enhancement degree coefficient according to the contrast feature value and the texture confidence; obtain the adaptive contrast limit value according to the enhancement degree coefficient and the number feature of the pixel points with different gray values in the road condition area;

[0016] Enhance the road condition area according to the adaptive contrast limit value through the limited contrast adaptive histogram equalization algorithm to obtain an enhanced road condition area; obtain an enhanced road condition image according to all the enhanced road condition areas.

[0017] Further, the step of obtaining different gray ranges according to the change feature of the number of pixel points in the gray histogram of the road condition area includes:

[0018] Perform curve fitting on the number of pixel points corresponding to all gray levels in the gray histogram to obtain a pixel point number representation curve; construct the gray range according to the gray values of the adjacent inflection points before and after any peak point in the pixel point number representation curve.

[0019] Further, the step of obtaining the gray distribution feature value according to the number difference feature of the pixel points with different grays in the gray range includes:

[0020] Calculate the excess kurtosis of the number of pixel points in the gray range and perform a positive correlation mapping to obtain the gray distribution feature value of the gray range.

[0021] Further, the step of obtaining the contrast feature value according to the gray feature of the road condition area, the number feature of the gray range, and the gray distribution feature value includes:

[0022] Calculate the average value of the gray distribution feature values of all gray ranges in the road condition area to obtain the gray distribution average value; calculate the product of the number of gray value types in the road condition area and the number of gray ranges to obtain the first factor; calculate the product of the reciprocal of the gray distribution average value and the first factor to obtain the contrast feature value of the road condition area.

[0023] Further, the step of clustering according to the gradient features of the pixel points in the road condition area to obtain two pixel point clusters includes:

[0024] Cluster the pixel points according to the gradient values of the pixel points in the road condition area through the K-means clustering algorithm to obtain two pixel point clusters.

[0025] Further, the step of obtaining the texture confidence according to the gradient difference feature of the pixel cluster and the pixel distance feature in the pixel cluster includes:

[0026] Calculate the average value of the pixel gradient values in the pixel cluster to obtain the average gradient value; calculate the average value of the Euclidean distances between all pixels and the nearest pixel in the pixel cluster corresponding to the maximum value of the average gradient value and perform a negative correlation mapping to obtain the adjacent feature value; calculate the product of the absolute value of the difference between the average gradient values of the pixel cluster and the adjacent feature value to obtain the texture confidence of the road condition area.

[0027] Further, the step of obtaining the enhancement degree coefficient according to the contrast feature value and the texture confidence includes:

[0028] Calculate the product of the reciprocal of the contrast feature value and the texture confidence to obtain the enhancement degree coefficient of the road condition area.

[0029] Further, the step of obtaining the adaptive contrast limit value according to the enhancement degree coefficient and the number feature of pixels with different gray values in the road condition area includes:

[0030] Calculate the ratio of the number of pixels in the road condition area to the maximum value of the number of pixels in the gray histogram to obtain the pixel number difference value; calculate the product of the pixel number difference value and the enhancement degree coefficient and normalize it to obtain the adaptive contrast limit value of the road condition area.

[0031] The beneficial effects of the embodiments of the present invention compared with the prior art are:

[0032] The road condition image acquired by the image acquisition device is enhanced by the processor, so that the contrast of the enhanced road condition image is higher and the road condition features are more obvious; the enhanced road condition image is transmitted to the display device, which can improve the driver's road condition recognition accuracy and driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the embodiments or the description of the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0034] Figure 1 is a system block diagram of a vision recognition system for vehicle assisted driving provided by Embodiment 1 of the present invention;

[0035] Figure 2It is a layout diagram of the installation position of a visual recognition display device for vehicle assisted driving provided in Embodiment 1 of the present invention. Detailed implementation manners

[0036] The embodiments of the present disclosure will be described in detail below. Examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present disclosure, and should not be construed as limiting the present disclosure.

[0037] It should be noted that the terms "first", "second", etc. in the description of the present disclosure and the above accompanying drawings are used to distinguish similar objects and do not necessarily have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the present disclosure. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present disclosure.

[0038] To illustrate the technical solution of the present invention, specific embodiments will be used for illustration below.

[0039] Refer to Figure 1 , which is a system block diagram of a visual recognition system for vehicle assisted driving provided in Embodiment 1 of the present invention. As Figure 1 shown, the system includes:

[0040] An image acquisition device S1, which is arranged outside the vehicle and is used to acquire the road condition images captured by the vehicle in real time.

[0041] During the driving of the vehicle, the road condition images are captured in real time by the image acquisition device installed outside the vehicle to obtain the road condition images, where the image acquisition device includes but is not limited to a camera; Figure 2 shows a layout diagram of the installation position of a visual recognition display device for vehicle assisted driving. Figure 2 The part pointed to by the number 2 in

[0042] is a camera outside the vehicle. This schematic diagram is only used to illustrate the installation position of the camera. The actual camera is not limited to be installed at the position shown in the figure. Similarly, it can also be installed at positions such as the vehicle hood, vehicle headlights, vehicle top, vehicle bumper, etc. according to different actual vehicle models. Since the clarity of the images captured in bad weather is relatively low, it is easy to affect the judgment of the driver and the vehicle system and reduce driving safety; therefore, it is necessary to enhance the captured images to improve the clarity and contrast.

[0043] The road condition image acquired by the image acquisition device is transmitted to a processor, and the road condition image is subjected to image enhancement processing in the processor. Currently, the contrast-limited adaptive histogram equalization algorithm is usually used to enhance the image and improve the image contrast. During the image enhancement process of this algorithm, in order to avoid over-enhancement of the image, a contrast limit value parameter is set to impose restrictions during local enhancement. However, during the driving of an automobile, due to the influence of adverse weather and other conditions, the clarity of images acquired at different times and positions may be affected to varying degrees. If a fixed contrast limit value parameter is used, some images may be over-enhanced or the enhancement effect may be poor, affecting the image quality and reducing the accuracy of image information recognition. Therefore, it is necessary to improve the contrast-limited adaptive histogram equalization algorithm to improve the image enhancement quality.

[0044] Furthermore, contrast-limited adaptive histogram equalization divides an image into blocks for local image enhancement to improve the enhancement quality. Therefore, the road condition image is divided into blocks to obtain different road condition regions. In the embodiments of the present invention, the default 8×8 block division method of this algorithm is used for block division to obtain different road condition regions. Since the display contrast of each road condition region may be different, the selection of the contrast limit value for enhancement of each road condition region may also be different. When the display contrast in the road condition region is low, a larger contrast limit value is required to improve the enhancement effect; if the display contrast in the road condition region is high, a smaller contrast limit value is required to avoid over-enhancement. Furthermore, the contrast situation of the road condition region can be analyzed based on the gray histogram of the road condition region. When the number of pixel points on both sides of the peak point in the gray histogram shows a normal distribution, it indicates that the gray values of most pixel points in the road condition region are close, lacking rich hierarchical changes and having a low contrast. Therefore, different gray ranges are obtained based on the change characteristics of the number of pixel points in the gray histogram of the road condition region.

[0045] Preferably, in the embodiments of the present invention, the steps of obtaining different gray ranges include: performing curve fitting on the number of pixel points corresponding to all gray levels in the gray histogram to obtain a pixel point number representation curve; the pixel point number representation curve represents the approximate number of pixel points corresponding to different gray levels in the road condition region, weakening the influence of the small difference in the number of pixel points corresponding to some adjacent gray values in the gray histogram on the selection of the gray range. Constructing a gray range based on the gray values of the front and rear adjacent inflection points of any peak point in the pixel point number representation curve. The gray range is used for subsequent analysis of the distribution characteristics of the number of pixel points at different gray levels, and each peak point in the pixel point number representation curve corresponds to a gray range.

[0046] Furthermore, when the number of pixel points corresponding to different gray levels in the gray range shows more normal distribution characteristics, it means that a large number of pixel points are more concentrated in similar gray levels, and the contrast is smaller; when the number of pixel points corresponding to different gray levels is more similar, the contrast is larger. Therefore, the gray distribution characteristic value is obtained according to the difference characteristics of the number of pixel points with different grays in the gray range; preferably, in the embodiment of the present invention, the steps of obtaining the gray distribution characteristic value include: calculating the excess kurtosis of the number of pixel points in the gray range and performing a positive correlation mapping to obtain the gray distribution characteristic value of the gray range. It should be noted that the excess kurtosis belongs to the prior art. When the excess kurtosis is a constant 0, it means that the distribution characteristic of the number of pixel points in the gray range shows a normal distribution. If the excess kurtosis is larger, it means that the shape of the peak is sharper, and the gray values of more pixel points are the same, and the contrast of the road condition image is smaller; if the excess peak value is less than 0, it means that the shape of the peak is flatter, and the number of pixel points corresponding to different grays is more similar, and the contrast of the road condition image is larger. Therefore, the larger the gray distribution characteristic value of the gray range, the lower the contrast.

[0047] When the types of gray values of the pixel points in the road condition area are fewer, it means that the gray distribution range of the pixel points in the road condition area is more concentrated, and the contrast is lower; if the number of gray ranges obtained in the road condition area is fewer, it means that the number of peak points is fewer, which means that the gray values of the pixel points in the road condition area are more concentrated, and the contrast is lower; if the gray distribution characteristic values of the gray ranges in the road condition area are all larger, the contrast of the road condition area is lower. Furthermore, the contrast characteristic value is obtained according to the gray characteristic, the number characteristic of the gray range, and the gray distribution characteristic value of the road condition area; preferably, in the embodiment of the present invention, the steps of obtaining the contrast characteristic value include: calculating the average value of the gray distribution characteristic values of all gray ranges in the road condition area to obtain the gray distribution average value; when the gray distribution average value is smaller, it means that the contrast of the road condition area is higher. Calculate the product of the number of types of gray values and the number of gray ranges in the road condition area to obtain the first factor; when the first factor is larger, it means that the number of types of gray values and the number of gray ranges in the road condition area are more, and the contrast of the road condition area is higher. Calculate the product of the reciprocal of the gray distribution average value and the first factor to obtain the contrast characteristic value of the road condition area. When the contrast characteristic value is larger, it means that the contrast of the road condition area is higher, and the contrast limit value should be smaller to avoid over-enhancement; when the contrast characteristic value is smaller, it means that the contrast of the road condition area is lower, and the contrast limit value should be larger to improve the enhancement degree. The formula for obtaining the contrast characteristic value includes:

[0048]

[0049] Wherein, D represents the contrast eigenvalue of the road condition area, G represents the average value of the gray-scale distribution, R represents the number of gray-scale value types in the road condition area, H represents the number of gray-scale ranges, and R*H represents the first factor.

[0050] Furthermore, during the driving of the vehicle, the camera will capture a large number of sky background areas. When the road condition area shows a pure-color sky or cloud, the obtained contrast eigenvalue is small, but there is no effective information in such images. If a large contrast limit value is used, it will lead to over-enhancement, highlighting the noise characteristics and reducing the image quality. When the effective information in the road condition area is affected by clouds or light intensity, the obtained contrast eigenvalue is also small, but a greater degree of enhancement is required to highlight the effective information. There are no edge textures in pure-color areas such as the sky background, while there are obvious edge textures in the road condition areas disturbed by clouds and light intensity. Therefore, clustering is performed according to the gradient characteristics of the pixel points in the road condition area to obtain two pixel point clustering clusters; preferably, in the embodiment of the present invention, the steps of obtaining two pixel point clustering clusters include: clustering the pixel points according to the gradient values of the pixel points in the road condition area through the K-means clustering algorithm to obtain two pixel point clustering clusters. It should be noted that the K-means clustering algorithm belongs to the prior art, and the specific steps will not be elaborated.

[0051] Furthermore, if there is an obvious difference in the gradient values of the two pixel point clustering clusters, it means that there are background pure-color areas and near-view road condition information in the road condition area; the gradient value of the background pure-color area is small, and the gradient value of the near-view road condition information is large due to the edge texture. If the pixel points in the pixel point clustering cluster with a large gradient value are closer, it can better indicate continuous edge textures rather than noise pixel points, and further indicate that there is effective information in the road condition area. Therefore, the texture confidence is obtained according to the gradient difference characteristics of the pixel point clustering clusters and the pixel point distance characteristics in the pixel point clustering clusters.

[0052] Preferably, in the embodiments of the present invention, the step of obtaining the texture confidence includes: calculating the average value of the gradient values of the pixel points in the pixel point clustering cluster to obtain the average gradient value. Calculating the average value of the Euclidean distances between all pixel points and the nearest pixel points in the pixel point clustering cluster corresponding to the maximum value of the average gradient value and performing a negative correlation mapping to obtain the adjacent feature value; when the adjacent feature value is larger, it means that the distances between the pixel points in the pixel point clustering cluster are closer, then it is more likely to be an edge texture, and it can better illustrate the existence of near-view road condition information in the road condition area. Calculating the product of the absolute value of the difference between the average gradient values of the pixel point clustering cluster and the adjacent feature value to obtain the texture confidence of the road condition area; when the absolute value of the difference between the average gradient values of two pixel point clustering clusters is smaller, it means that the road condition area is more likely to be a solid color background area, and the gradient values of the pixel points are similar; on the contrary, when the absolute value of the difference between the average gradient values is larger, it means that the possibility of representing different road condition obstacles in the road condition area is greater, there are more edge textures, and the larger the adjacent feature value, the more it can further illustrate the existence of edge textures. Therefore, the larger the texture confidence, the more likely it is that there is effective road condition information in the road condition area, and the more the enhancement degree should be increased; when the texture confidence is smaller, it means that the possibility of the existence of effective road condition information in the road condition area is smaller, and the enhancement degree should not be increased.

[0053] Furthermore, the enhancement degree coefficient can be obtained according to the contrast feature value and the texture confidence; preferably, in the embodiments of the present invention, the step of obtaining the enhancement degree coefficient includes: calculating the product of the reciprocal of the contrast feature value and the texture confidence to obtain the enhancement degree coefficient of the road condition area. When the enhancement degree coefficient is larger, a larger contrast limit value is required to increase the enhancement degree and highlight the contrast of the road condition area.

[0054] Furthermore, if the value of the maximum peak point in the grayscale histogram of the road condition area is larger and the number of pixel points is more, in order to avoid excessive enhancement of this area resulting in contrast imbalance or noise amplification, the contrast limit value should be appropriately reduced. Reducing this contrast limit value parameter can limit the height of the peak in the grayscale histogram, clip the gray levels with too many pixel points and redistribute them to other gray levels, thereby smoothing the histogram distribution, and further reducing the problem of overly strong local contrast in over-bright or over-dark areas, making the equalization effect more natural, and at the same time preventing detail loss or enhancing noise. Therefore, an adaptive contrast limit value is obtained according to the enhancement degree coefficient and the characteristics of the number of pixel points with different gray values in the road condition area; preferably, in the embodiment of the present invention, the steps of obtaining the adaptive contrast limit value include: calculating the ratio of the number of pixel points in the road condition area to the maximum value of the number of pixel points in the grayscale histogram to obtain the pixel point number difference value; the larger the pixel point number difference value, it means that the value of the maximum peak point in this grayscale histogram is larger and the number of pixel points is more, then the contrast limit value should be appropriately reduced. Calculate the product of the pixel point number difference value and the enhancement degree coefficient and normalize it to obtain the adaptive contrast limit value of this road condition area; in the embodiment of the present invention, the normalization function is y represents the adaptive contrast limit value, e represents the natural constant, k is 2.9 in the embodiment of the present invention, b is 1.4, and x is the independent variable to be normalized. The larger the adaptive contrast limit value, it means that a greater degree of enhancement is required to highlight the effective information in the road condition area; conversely, the smaller the adaptive contrast limit value, it means that the contrast of this road condition area is larger, or it is a pure color area without effective information, and no excessive enhancement is required.

[0055] After obtaining the adaptive contrast limit value of each road condition area, the road condition area can be enhanced by the contrast-limited adaptive histogram equalization algorithm according to the adaptive contrast limit value to obtain the enhanced road condition area; it should be noted that the contrast-limited adaptive histogram equalization algorithm belongs to the prior art, and the specific steps will not be elaborated; the enhanced road condition area obtains the adaptive contrast limit value according to its own image information characteristics, making the enhancement degree more in line with the image characteristics, and can effectively improve the enhancement quality compared with the fixed contrast limit value parameter. Furthermore, an enhanced road condition image is obtained based on all the enhanced road condition areas. The enhanced road condition image can significantly enhance the contrast of different areas compared with the captured road condition image, enabling the driver to obtain more accurate effective information.

[0056] The display device S3, which is arranged on the vehicle instrument panel and faces the driver, is used to receive and display the enhanced road condition image to assist the driver in driving.

[0057] Transmit the enhanced road condition image to the display device on the vehicle interior instrument panel and facing the driver, Figure 2The part pointed by the number 1 is the display device on the dashboard inside the vehicle. This figure also only shows one situation where the display device is installed inside the car. In practice, the display device can also be installed in the rearview mirror, the central control display screen, the left front side of the driver, etc. according to different car models. During driving, the driver can more accurately judge the road condition through the enhanced road condition image, thereby improving the driver's road condition recognition accuracy and driving safety.

[0058] In order to achieve the long-term development and wider application of this technical solution, we will continue to optimize and expand the intelligent scene detection function on the existing basis to further improve the adaptability and safety of the system in complex driving environments. Specifically, the overall technical solution will integrate more intelligent environmental perception capabilities, combined with deep learning and continuous training of large-scale environmental data, so that the system can intelligently identify and classify a variety of different driving scenarios, especially in bad weather, strong light changes and complex road conditions. Through this optimization, the system can automatically adjust the image processing strategy to more effectively deal with the interference of various environmental factors on visual information.

[0059] The intelligent detection and classification system will be able to conduct real-time analysis of a variety of driving environments, including but not limited to different weather conditions (such as heavy rain, snow, haze, sandstorms, etc.), strong light changes (such as direct sunlight, sunset, or rapid changes in light at tunnel entrances and exits), and dynamic monitoring of objects outside the vehicle (such as pedestrians, other vehicles, obstacles, etc.). The system will use advanced object detection and classification technology to accurately identify these external factors and provide optimized visual feedback to the driver in a timely manner, thereby improving the driver's perception of complex environments.

[0060] This process will continuously optimize the image processing algorithm through continuous training of deep learning models and large amounts of environmental data, and improve its adaptability and recognition accuracy to different scenarios. With the iteration of technology, the system will be able to more intelligently identify various environmental changes and potential risks, automatically adjust image output, and ensure that drivers can always obtain clear and accurate road information, whether in extreme weather, complex lighting conditions, or in traffic scenes with dense traffic.

[0061] In addition, the intelligent detection and classification capabilities of the technical solution of the present invention will further expand the application scenarios. In addition to conventional driving environments, the system will also demonstrate strong adaptability in night driving, extreme weather conditions, complex road conditions (such as highways, bridges, tunnels, etc.) and multi-lane traffic scenarios. Through real-time environmental perception and image enhancement, the system will be able to effectively reduce the driver's visual burden, improve reaction speed, and significantly reduce misjudgment caused by environmental interference, thereby further enhancing driving safety.

[0062] To ensure the continuous optimization and long-term stability of the system, continuous innovation will be carried out in data collection and processing capabilities in the future. By continuously enhancing the deep learning training of multi-dimensional environmental data, the intelligence level of the system will be further improved. With the evolution of technology, the algorithms of the system will continue to evolve, enabling it to handle more complex environmental changes and driving scenarios, and gradually moving towards a fully adaptive intelligent assisted driving system.

[0063] In summary, the technical solution of the present invention not only meets the actual needs in the current environment, but also maintains foresight in technological development. It is planned to comprehensively improve the accuracy, intelligence and safety of the driving assistance system by continuously optimizing intelligent detection, object classification and image processing algorithms. The goal of this application is to provide a more accurate, efficient and reliable assisted driving experience for drivers, while promoting the application of intelligent driving technology in more scenarios, and ultimately achieving broader traffic safety protection.

[0064] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A visual recognition system for automobile assisted driving, characterized in that: The visual recognition system comprises: An image acquisition device, arranged outside the vehicle, for acquiring road condition images taken by the vehicle in real time; a processor connected to the image acquisition device and the display device, configured to receive the road condition image sent by the image acquisition device, perform image enhancement on the road condition image to obtain an enhanced road condition image, and send the enhanced road condition image to the display device; The display device is arranged on the dashboard of the car and faces the driver, and is used for receiving and displaying the enhanced road condition image to assist the driver in driving.

2. A visual recognition system for automobile assisted driving according to claim 1, characterized in that: The step of performing image enhancement on the road condition image to obtain an enhanced road condition image includes: Dividing the traffic condition image into blocks to obtain different traffic condition areas; Different grayscale ranges are obtained according to the characteristics of the number of pixel points in the grayscale histogram of the road condition area; grayscale distribution characteristic values ​​are obtained according to the characteristics of the number of pixel points of different grayscales in the grayscale range; contrast characteristic values ​​are obtained according to the grayscale characteristics of the road condition area, the number characteristics of the grayscale range, and the grayscale distribution characteristic values; Clustering is performed according to the gradient characteristics of the pixels in the road condition area to obtain two pixel clusters; texture confidence is obtained according to the gradient difference characteristics of the pixel clusters and the distance characteristics of the pixels in the pixel clusters; an enhancement degree coefficient is obtained according to the contrast feature value and the texture confidence; an adaptive contrast limit value is obtained according to the enhancement degree coefficient and the number of pixels with different gray values ​​in the road condition area; According to the adaptive contrast limit value, the road condition area is enhanced by a contrast-limited adaptive histogram equalization algorithm to obtain an enhanced road condition area; and an enhanced road condition image is obtained according to all enhanced road condition areas.

3. A visual recognition system for automobile assisted driving according to claim 2, characterized in that: The step of obtaining different grayscale ranges according to the change characteristics of the number of pixels in the grayscale histogram of the road condition area comprises: Curve fitting is performed according to the number of pixels corresponding to all gray levels in the grayscale histogram to obtain a pixel number characterization curve; and the grayscale range is constructed according to the grayscale values ​​of the adjacent inflection points before and after any peak point in the pixel number characterization curve.

4. The visual recognition system for automobile assisted driving according to claim 2, characterized in that: The step of obtaining the grayscale distribution characteristic value according to the difference characteristics of the number of pixels of different grayscales in the grayscale range comprises: The excess kurtosis of the number of pixels in the grayscale range is calculated and positively correlated to obtain the grayscale distribution characteristic value of the grayscale range.

5. The visual recognition system for automobile assisted driving according to claim 2, characterized in that: The step of obtaining a contrast characteristic value according to the grayscale characteristics of the road condition area, the quantity characteristics of the grayscale range, and the grayscale distribution characteristic value comprises: Calculate the average value of the grayscale distribution characteristic values ​​of all grayscale ranges in the road condition area to obtain the grayscale distribution average value; calculate the product of the number of grayscale value types in the road condition area and the number of grayscale ranges to obtain a first factor; calculate the product of the inverse of the grayscale distribution average value and the first factor to obtain the contrast characteristic value of the road condition area.

6. The visual recognition system for automobile assisted driving according to claim 2, characterized in that: The step of clustering the pixels in the road condition area according to their gradient features to obtain two pixel clusters comprises: The pixels are clustered according to the gradient values ​​of the pixels in the road condition area using a K-means clustering algorithm to obtain two pixel clusters.

7. The visual recognition system for automobile assisted driving according to claim 2, characterized in that: The step of obtaining texture confidence according to the gradient difference feature of the pixel point clustering cluster and the pixel point distance feature in the pixel point clustering cluster comprises: The average value of the gradient values ​​of the pixels in the pixel cluster is calculated to obtain the average gradient value; in the pixel cluster corresponding to the maximum value of the average gradient value, the average value of the Euclidean distances between all pixels and the nearest pixel is calculated and negatively correlated to obtain adjacent eigenvalues; the product of the absolute value of the difference between the average gradient values ​​of the pixel cluster and the adjacent eigenvalue is calculated to obtain the texture confidence of the road condition area.

8. The visual recognition system for automobile assisted driving according to claim 2, characterized in that: The step of obtaining the enhancement degree coefficient according to the contrast feature value and the texture confidence comprises: The product of the inverse of the contrast characteristic value and the texture confidence is calculated to obtain an enhancement degree coefficient of the road condition area.

9. The visual recognition system for automobile assisted driving according to claim 2, characterized in that: The step of obtaining an adaptive contrast limit value according to the enhancement degree coefficient and the number characteristics of pixels with different gray values ​​in the road condition area comprises: The ratio of the number of pixels in the road condition area to the maximum value of the number of pixels in the grayscale histogram is calculated to obtain a pixel number difference value; the product of the pixel number difference value and the enhancement degree coefficient is calculated and normalized to obtain an adaptive contrast limit value of the road condition area.

Citation Information

Patent Citations

  • Lane line detection method based on machine vision under strong illumination condition

    CN107895151A

  • Image processing method, equipment and device

    CN115100048A

  • Traffic engineering road condition visual detection method and system

    CN118015002A

  • Electronic guideboard based on visual unit structure and road condition target self-identification method thereof

    CN118135506A

  • Automatic identification method and system for bill shot image

    CN118334674A

Cited By

  • Signal anti-interference method and device for satellite navigation signal receiver

    CN121500341A