With the rapid development of road traffic and the increasingly explosion in auto inventory, problems concerning traffic mobility and safety, have become more serious in most developed countries in recent years. Statistics show that most of traffic accidents are caused by drivers drowsiness, inatten...
With the rapid development of road traffic and the increasingly explosion in auto inventory, problems concerning traffic mobility and safety, have become more serious in most developed countries in recent years. Statistics show that most of traffic accidents are caused by drivers drowsiness, inattentiveness. Autonomous driving system is the design used to be warning the driver when the vehicle leaving the lane or a collision, so it can help to avoid traffic accidents. Lane detection based on computer vision is a key technology in autonomous driving system of intelligent vehicles. Although many researchers have made great efforts on the research of this field, the accurate detection of road is still faced with many problems because of the variety and complexity of the environment , such as shadows, illumination variance, pavement dilapidation, vehicle occlusion and background interference. In order to overcome these above problems, in this thesis deep research is carried out for structured road to realize reliable lane detection under complex background. The main contents of this thesis are as follows: Firstly, the thesis introduces the background and significance about the lane detection based on edge detection, and then the research status of the thesis topic is reviewed. Finally, it makes a brief analysis. Secondly, this paper describes the tools required for programming-MATLAB, and makes a brief introduction to the two features of MATLAB. Thirdly, the thesis describes the technology related to image pre-processing module such as image gray, image filtering, edge detection. Experiments was performed for the selected methods. Fourthly, making a brief introduction of principle of Hough transform and using Hough to extract the lane marker. A lane detection algorithm based on linear-hyperbola model is proposed, which is to overcome the difficulties that simple road model cannot descript different shapes of road, and complex road model is vulnerable to be disturbed by background. Based on the FIR filter, OTUS method is adopted to solve the problem that traditional edge detection is liable to be influenced by high brightness area. and finally the proposed algorithm can complete reliably structured lane detection of different shapes. Finally, conclusions are given with recommendation for future work. Overall, in this paper, the proposed algorithm can effectively extract the lane, with better robustness, as compared to some existing main algorithms. Experimental results show that the proposed methods in this thesis can achieve reliable and effective lane detection in many complicated situations.
With the rapid development of road traffic and the increasingly explosion in auto inventory, problems concerning traffic mobility and safety, have become more serious in most developed countries in recent years. Statistics show that most of traffic accidents are caused by drivers drowsiness, inattentiveness. Autonomous driving system is the design used to be warning the driver when the vehicle leaving the lane or a collision, so it can help to avoid traffic accidents. Lane detection based on computer vision is a key technology in autonomous driving system of intelligent vehicles. Although many researchers have made great efforts on the research of this field, the accurate detection of road is still faced with many problems because of the variety and complexity of the environment , such as shadows, illumination variance, pavement dilapidation, vehicle occlusion and background interference. In order to overcome these above problems, in this thesis deep research is carried out for structured road to realize reliable lane detection under complex background. The main contents of this thesis are as follows: Firstly, the thesis introduces the background and significance about the lane detection based on edge detection, and then the research status of the thesis topic is reviewed. Finally, it makes a brief analysis. Secondly, this paper describes the tools required for programming-MATLAB, and makes a brief introduction to the two features of MATLAB. Thirdly, the thesis describes the technology related to image pre-processing module such as image gray, image filtering, edge detection. Experiments was performed for the selected methods. Fourthly, making a brief introduction of principle of Hough transform and using Hough to extract the lane marker. A lane detection algorithm based on linear-hyperbola model is proposed, which is to overcome the difficulties that simple road model cannot descript different shapes of road, and complex road model is vulnerable to be disturbed by background. Based on the FIR filter, OTUS method is adopted to solve the problem that traditional edge detection is liable to be influenced by high brightness area. and finally the proposed algorithm can complete reliably structured lane detection of different shapes. Finally, conclusions are given with recommendation for future work. Overall, in this paper, the proposed algorithm can effectively extract the lane, with better robustness, as compared to some existing main algorithms. Experimental results show that the proposed methods in this thesis can achieve reliable and effective lane detection in many complicated situations.
주제어
#차선검출 모델기반 영상처리
※ AI-Helper는 부적절한 답변을 할 수 있습니다.