Road condition detection system

Abstract

The utility model relates to a road condition detection system, which comprises a high-position camera, a low-position camera and a vehicle-mounted computer, the high-position camera and the low-position camera are respectively arranged at positions, capable of forming different vertical vision with a road, on an automobile, and are used for synchronously shooting a detection target on the road surface, respectively obtaining a first image and a second image of the detection target and transmitting the first image and the second image to the vehicle-mounted computer, and the vehicle-mounted computer is used for comparing and analyzing the first image and the second image; the road condition is detected through the high-position camera, the low-position camera and the vehicle-mounted computer, and when a pit, an obstacle or a broken road which endanger driving safety is detected, the vehicle-mounted computer brakes or avoids a vehicle, so that the driving safety is ensured. According to the system, road condition detection is realized by comparing different image features obtained by comparing different vertical sight angles, so that the road defect detection level of a current vehicle-mounted environment sensing system can be effectively improved, a vehicle is braked or avoided in time, and more reliable support is provided for automatic driving and auxiliary driving technologies.

Description

TECHNICAL FIELD

[0001] The utility model relates to vehicle-mounted environment perception technical field, concretely is a road condition detection system. BACKGROUND

[0002] With the rapid development of the automobile industry, intelligent driving is increasingly valued. In the process of driving, the complexity of the road environment poses many challenges to the safety of intelligent driving. Among them, roadblocks and potholes are common safety hazards, which can cause vehicle damage and even traffic accidents.

[0003] At present, the video system for detecting roadblocks and potholes is horizontally arranged, while obstacles and potholes often need to be compared and calculated vertically. The vertically arranged video system has many advantages over the horizontally arranged video system in comparison and calculation.

[0004] In real life, people confirm whether the pattern on the road ahead is an obstacle or a pothole by walking a few steps to watch the pattern change, or by lowering their heads or standing on their toes. This shows that the best way to judge whether the obstacle or pothole in front is real or an illusion of a flat pattern is to look at it from different vertical angles. It is also feasible to use the above method to determine whether the road ahead is a broken road.

[0005] Therefore, it is a demand for people to develop a road condition detection system with good economy and simple analysis process. SUMMARY

[0006] In view of the problems existing in the prior art, the purpose of the utility model is to provide a road condition detection system that can accurately obtain various road condition information.

[0007] In order to achieve the above purpose, the utility model adopts the following technical scheme:

[0008] A road condition detection system, comprising a high-position camera, a low-position camera and a vehicle-mounted computer;

[0009] The high-position camera and the low-position camera are respectively installed on the vehicle at positions that can form different vertical views with the road, for synchronously shooting the detection target on the road surface, obtaining the first image and the second image of the detection target respectively and transmitting them to the vehicle-mounted computer, and the vehicle-mounted computer is used for comparing and analyzing the first image and the second image; through the system, obstacles, potholes or broken roads can be found in time, and the vehicle can be braked or avoided; wherein,

[0010] When the vehicle-mounted computer detects that the ratio of the longitudinal length of the detection target in the first image to the height H2 of the second image is not equal to the H1 / H2 of the surrounding road pattern, it is judged that the detection target is a roadblock or a pothole;

[0011] When a certain pattern in the first image is not shown at a corresponding position in the second image, and the H1 / H2 ratio of the pattern at or below the position in the second image is not equal to the H1 / H2 of the surrounding road pattern, it is judged that the corresponding position in the second image is blocked by a protruding object on the detected target, and the protruding object is the upper part of the roadblock;

[0012] When the first feature in the first image is not shown at a corresponding position in the second image, and the H1 / H2 ratio of the pattern at or below the position in the second image is equal to the H1 / H2 of the surrounding road pattern, it is judged that the first feature is a pit;

[0013] When the detected object is a front road, and the first longitudinal length of the front road in the first image is much greater than the second longitudinal length in the second image, it is judged that the detected object is a downhill road;

[0014] When the detected object is a front road, and the first longitudinal length of the front road in the first image is equal to the second longitudinal length in the second image, it is judged that the detected object is a broken road.

[0015] Further, the high-position camera is installed on the top of the car, and the low-position camera is installed on the front of the car.

[0016] Further, the angle between the line of sight of the high-position camera and the road surface is greater than the angle between the line of sight of the low-position camera and the road surface.

[0017] Overall, the present application has the following advantages:

[0018] The present application simulates the behavior mode of human observing objects from different angles to judge their authenticity, and arranges high-position cameras, low-position cameras and vehicle-mounted computers on the car, so that obstacles, pits or broken roads in the road can be more reliably found, the detection level of the vehicle-mounted environmental perception system on the road condition can be effectively improved, and more reliable support is provided for automatic driving and auxiliary driving technology. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 It is a schematic diagram for detecting a roadblock by using a road condition detection system.

[0020] Figure 2 It is a comparative schematic diagram of the longitudinal length of the roadblock in the first image and the second image.

[0021] Figure 3 It is a schematic diagram for detecting a pit by using a road condition detection system.

[0022] Figure 4 It is a schematic diagram for detecting a downhill road by using a road condition detection system.

[0023] Figure 5This is a schematic diagram illustrating the use of a road condition detection system to detect dead-end roads.

[0024] In the picture:

[0025] 11 - High-position camera; 12 - Low-position camera;

[0026] 21-Roadblock, 22-Pit, 23-Downhill road, 24-Dead end of road;

[0027] 31 - First feature, 32 - Second feature. Detailed Implementation

[0028] The present invention will now be described in further detail.

[0029] like Figure 1 , Figures 3-5 As shown, a road condition detection system includes a high-position camera 11, a low-position camera 12, and an on-board computer.

[0030] The high-position camera 11 and the low-position camera 12 are respectively installed at different heights on the central axis of the car. Preferably, the high-position camera 11 is installed on the top of the car and the low-position camera 12 is installed at the front of the car. The high-position camera 11 is positioned further back than the low-position camera 12.

[0031] The high-position camera 11 and the low-position camera 12 are used to simultaneously capture the detection target on the road surface, respectively obtain the first image and the second image of the detection target, and transmit them to the vehicle computer.

[0032] The onboard computer receives the first and second images, compares and analyzes them, and determines the road conditions based on the comparison and analysis results, including analyzing whether the detected target is a road obstacle 21, a pit 22, a downhill road 23, or a dead end 24.

[0033] In one embodiment, such as Figure 1 , Figure 2 , Figure 3 As shown, the comparative analysis includes the onboard computer acquiring the first longitudinal length H1 of the detected target in the first image (e.g., ...). Figure 2 (as shown in the left image) and the second vertical length H2 in the second image (as shown in the left image) Figure 2 (As shown in the image on the right) When the ratio between the first longitudinal length H1 and the second longitudinal length H2 exceeds the reference value range, the target is determined to be a roadblock 21 or a pit 22. The reference value is obtained by simultaneously shooting planar targets on the road surface with the high-position camera 11 and the low-position camera 12. The vehicle computer calculates the H1 / H2 value for each horizontal target on the road, and then uses the normal distribution function to calculate the reference value range for that horizontal direction and stores it in the vehicle computer.

[0034] Specifically, because the longitudinal length of the planar target and the stereoscopic target imaged in the high camera 11 or the low camera 12 is different, the ratio of the longitudinal length of the planar target in the images taken by the high camera 11 and the low camera 12 is also different from the ratio of the longitudinal length of the stereoscopic target in the images taken by the high camera 11 and the low camera 12. Therefore, the vehicle-mounted computer calculates the H1 / H2 value of each horizontal target in the road, and then calculates the reference value range of the horizontal target by using a normal distribution function. When the measured target value exceeds the reference value range, it indicates that the detected target is a stereoscopic target, and it can be inferred that the detected target is the roadblock 21 or the pit 22.

[0035] When the high camera 11 can take the first feature 31 behind the roadblock 21 or inside the pit 22, the first image contains the image of the detected target and the image of the first feature 31 adjacent to the top of the image of the detected target, and sometimes also contains the image of the second feature 32 above the image of the first feature 31, as shown in the left image. Figure 2 Because of the obstruction of the roadblock 21 or the pit 22, the low camera 12 cannot take the first feature 31 behind the roadblock 21 or inside the pit 22, and can only take the second feature 32 behind the first feature 31 (in the case of the roadblock 21) or the second feature 32 above the first feature 31 (in the case of the pit 22), as shown in the right image. That is, in the second image, the image of the first feature 31 is missing between the image of the detected target and the image of the second feature 32. At this time, it can be inferred that the detected target is the roadblock 21 or the pit 22. Figure 2

[0036] In another embodiment, the comparison analysis includes that when a certain pattern in the first image is not displayed at the corresponding position in the second image, and the H1 / H2 ratio of the pattern at or below the corresponding position in the second image is not equal to the H1 / H2 of the surrounding road pattern, it is judged that the corresponding position in the second image is blocked by a protruding object on the detected target, and the protruding object is the upper part of the roadblock 21.

[0037] As shown in Figure 1 , Figure 2 When the first image contains the image of the detected target and the image of the first feature 31 adjacent to the top of the image of the detected target, the second image contains the image of the detected target and the image of the second feature 32 adjacent to the top of the image of the detected target, and the image of the first feature 31 is different from the image of the second feature 32, the first feature 31 is blocked by a certain protruding object on the detected target, and therefore the high camera 11 can take the first feature 31, while the low camera 12 cannot take the first feature 31. At this time, it is judged that the protruding object is the top part of the roadblock 21.

[0038] ​In another embodiment, the comparison analysis includes that when the first feature 31 in the first image is not shown at the corresponding position in the second image, and the H1 / H2 ratio of the pattern at or below the position in the second image is equal to the H1 / H2 of the surrounding road pattern, the first feature 31 is determined as the pit 22.

[0039] As shown in FIG. 6, when the first image contains the image of the detection target and the image of the first feature 31 above the detection target, the second image contains the image of the detection target and the image of the second feature 32 above the detection target, and the image of the first feature 31 is different from the image of the second feature 32, and the H1 / H2 ratio of the pattern at or below the corresponding position in the second image is equal to the H1 / H2 of the surrounding road pattern, it indicates that the detection target at the position is a flat road rather than a three-dimensional target, and the first feature 31 is determined as the pit 22. Figure 3

[0040] When the first feature 31 and the second feature 32 are similar, the prior art can be used for identification to improve the accuracy of distinguishing, which will not be described here.

[0041] In another embodiment, as shown in FIG. 7, the comparison analysis includes that when the first longitudinal length of the detection target in the first image is much greater than the second longitudinal length in the second image, the detection target is determined as the downhill. Figure 4

[0042] Specifically, due to the obstruction of the downhill road 23, the low-position camera 12 can only capture a small part of the front end of the downhill road 23, so the second longitudinal length in the second image is shorter; since the installation position of the high-position camera 11 is higher than that of the low-position camera 12, it can capture a longer distance of the downhill road 23, and the first longitudinal length in the first image is longer and usually much greater than the second longitudinal length in the second image. Therefore, the significant difference in the longitudinal length of the detection target in the first image and the second image can be used to infer that the detection target is the downhill road 23.

[0043] In another embodiment, as shown in FIG. 8, the comparison analysis includes that when the first longitudinal length of the detection target in the first image is equal to the second longitudinal length in the second image, the detection target is determined as the dead-end road 24. Figure 5

[0044] Specifically, when the front of the car is the dead-end road 24, the high-position camera 11 and the low-position camera 12 can only capture the end of the dead-end road 24, and the longitudinal length of the detection target in the first image and the second image is the same. Therefore, the first longitudinal length of the detection target in the first image is substantially equal to the second longitudinal length in the second image, which can be used to infer that the detection target is the dead-end road 24.

[0045] ​​​In actual driving scenarios, the detection of roadblocks 21, pits 22, downhill roads 23 or broken roads 24 often needs to be observed and analyzed from different vertical angles to accurately distinguish real roadblocks 21, pits 22, downhill roads 23 or broken roads 24 from the illusion of flat patterns. The utility model provides a double-camera image comparison device based on different vertical angles, which is used for detecting road conditions in a vehicle-mounted environmental perception system. By simulating the behavior pattern of humans observing objects from different angles to judge their authenticity, only high-position cameras 11, low-position cameras 12 and vehicle-mounted computers need to be arranged on the vehicle, and the rough detection of road conditions can be completed through simple analysis, which can effectively improve the detection of road conditions by the vehicle-mounted environmental perception system and provide more reliable support for automatic driving and auxiliary driving technology. Compared with arranging laser radars, the detection system provided by the utility model is more economical, and the detection method is more reliable.

[0046] The image comparison and judgment method disclosed by the utility model all adopt prior art.

[0047] The above-mentioned embodiments are preferred embodiments of the utility model, but the embodiments of the utility model are not limited by the above-mentioned embodiments, and any change, modification, replacement, combination, simplification made without departing from the spirit and principle of the utility model should be equivalent replacement methods, which are all included in the protection scope of the utility model.

Claims

1. A road condition detection system, characterized by: The system comprises a high-position camera, a low-position camera and a vehicle-mounted computer; The high-position camera and the low-position camera are respectively installed on the vehicle at positions capable of forming different vertical views with the road, and are used to synchronously capture a detection target on the road surface, to obtain a first image and a second image of the detection target respectively and transmit the first image and the second image to the vehicle-mounted computer, and the vehicle-mounted computer is used to compare and analyze the first image and the second image and determine the road condition according to the comparison and analysis result; thus, through the system of the high-position camera, the low-position camera and the vehicle-mounted computer, a pit, an obstacle or a dead-end road can be found in time and braking or avoidance can be implemented on the vehicle; wherein, When the vehicle-mounted computer detects that a ratio of a longitudinal length of the detection target in the first image to a longitudinal length of the detection target in the second image is not equal to H1 / H2 of a surrounding road pattern, it is determined that the detection target is an obstacle or a pit; When a pattern in the first image is not displayed at a corresponding position in the second image, and a ratio of H1 / H2 of the pattern at or below the corresponding position in the second image is not equal to H1 / H2 of the surrounding road pattern, it is determined that the corresponding position in the second image is blocked by a convex object on the detection target, and the convex object is an upper part of the obstacle; When a first feature in the first image is not displayed at a corresponding position in the second image, and a ratio of H1 / H2 of a pattern at or below the corresponding position in the second image is equal to H1 / H2 of the surrounding road pattern, it is determined that the first feature is a pit; When the detection target is a front road, and a first longitudinal length of the front road in the first image is much greater than a second longitudinal length of the front road in the second image, it is determined that the detection target is a downhill road; When the detection target is a front road, and the first longitudinal length of the front road in the first image is equal to the second longitudinal length of the front road in the second image, it is determined that the detection target is a dead-end road.

2. The road condition detection system of claim 1, wherein: The high-position camera is installed on a top of the vehicle, and the low-position camera is installed on a front of the vehicle.

3. The road condition detection system of claim 1, wherein: An angle between a line of sight of the high-position camera and the road surface is greater than an angle between a line of sight of the low-position camera and the road surface.

Images