The Study Of Triangle Traffic Sign Detection

The automobile is more important in the human being's life. How to manage and regulate the traffic problem caused by the automobile is a key subject. The development of automobile is intelligent. Traffic signs detection is one subject of intelligent automobile. It provides important information for guiding, warning, and regulating to the drivers. As one of the major tasks of drive assistance systems, traffic sign detection has attracted more and more attention from scientists. This thesis begins with an introduction to the development of traffic sign detection and follows a brief description of several primary methods for traffic sign detection. Especially some approaches for triangle traffic sign detection are introduced. The method begins with edge detection and then carries up a research into the use of fast symmetry and Hough transform in detection of the triangle traffic sign. The inner circle point of the triangle is detected by fast radial symmetry, then a specified window, which center point (it is also the inner circle point) is regarded as the original point in plane coordinates, is used to compute the Hough Transform of the small regions of the image. Peaks of the Hough image which correspond to the line segments are then extracted. A triangle is detected when the three lines satisfy the certain conditions. A comparison is made between the experiment result of this algorithm and the other algorithms which based on edge information. The experiment proves that the presented approach is robust and right-angled triangle, isosceles triangle objects also can be detected.