Check out RSS, or use RSS reader to subscribe this item
Confirmation
Authentication email has already been sent, please check your email box: and activate it as soon as possible.
You can login to My Profile and manage your email alerts.
Sponsored by the Center for Science and Technology Development of the Ministry of Education
Supervised by Ministry of Education of the People's Republic of China
This paper presents a new pedestrian detection method using a monochrome video camera for driving assistance system. The algorithm is based on the cascade classifier structure first proposed by Viola for face detection. Several extensions have been proposed to make the detection framework more competent for pedestrian detection. First, two new haar like features using two separated rectangles are proposed to enrich the feature pool. Second, the decision tree is used as weak hypothesis of Real Adaboost to capture the dependencies between these features and a new splitting criterion for the tree is used to minimize the error bound of Real Adaboost directly. Third, a tree cascade classifier structure is adopted to deal with intra-class difference of pedestrian. Finally, a kalman filter is used to combine single frame detections together. Pooling together all the strategies, a new pedestrian detection system has been developed which can detect pedestrians as small as 14×28 pixels at 15f/s for a 320×240 image on a P4 3.0GHz computer.
Keywords:Pedestrian Detection;Real Adaboost;Decision Tree