Implementation And Application Of Line-segment Detection Algorithm Based On Improved Hough Transform

With the development of economy and society,the popularization of cars has brought great convenience to people's life.However,at the same time,the nuber of casualties and accidents caused by drunk driving and fatigue driving also increases year by year.With the development of artificial intelligence technology,self-driving cars have entered people's lives,which will greatly reduce the traffic accident rate and bring better travel experience.As one of the key modules of the automatic driving system,lane line detection plays an important role in collision prevention,automatic parking,automatic parallel parallel road and other tasks.Using monocular camera to take road images and using computer vision and pattern recognition technology to detect lane lines is one of the hot research areas at present.Because the lane line is mainly shown as yellow or white line segment with obvious contrast with road in the image,it is of great significance to study the line segment detection algorithm for further research on lane line detection.In digital image,line segment has a series of property parameters such as length,width,position and direction.To extract line segment features in the image,it is necessary not only to detect line segment in the image,but also to calculate the line segment's attribute parameters.Lane line detection system needs to meet three requirements:First,when a large amount of noise is included in the image,error detection,leakage detection and calculation deviation of line segment parameters will be caused.The system needs to be able to effectively filter noise.Second,the line segment in the image has various attributes,such as length,width,position,direction,etc.A series of attribute parameters of the lane line are required when detecting the lane line.Third,the image often contains multiple parallel,crossing and col-linear lane lines.The system needs to be able to detect multiple line segments at the same time.In view of the above requirements,the main work of this paper is as follows:1)In-depth research and analysis of line segment detection method based on Hough transform was conducted.And finally improved line segment detection algorithm based on Hough transform was adopted as the core algorithm of lane line detection system.2)The improved Hough transform based line segment detection algorithm is implemented in Python language,which can be use for single-line segment detection and multiple line-segment detection.And the accuracy and efficiency of the algorithm is verified via using simulated data and natural scene data.