부산대학교 도서관

Typically, road lanes are solid or dashed line formations that are continuous on the surface. As the driving sceneries are continuous as well as substantially overlap, the placement of lanes in one frame is highly correlated with their position in the next frame. Computer vision-related machine learning algorithms have also advanced rapidly in recent years, becoming both more efficient & more effective as high-precision optical and electronic sensors become more commonplace, and real-time driving scene recognition is becoming more feasible. Recent years have seen several technical breakthroughs in the field of road safety, as the number of accidents has risen at an alarming pace, with driver inattention being one of the primary causes. To minimize the incidence of accidents as well as maintain safety, technological advances are required. Lane Detection Systems, which operate to recognize the lane boundaries on the road as well as alert the driver if he changes and goes to incorrect lane markings, are one method of achieving this goal. A lane detection system is a crucial element of many technologically advanced transportation systems. This research uses the Hough Transform technique for lane identification.