A proactive crash mitigation system is proposed to enhance crash avoidance and survivability of intelligent vehicles. Accurate object detection and recognition is a prerequisite for such a system, as system component deployment algorithms rely on accurate hazard detection, recognition, and tracking information. We present a vision-based approach to detect and recognize vehicles and traffic signs, obtain information, and track multiple objects by using a sequence of color images taken from a moving vehicle. The system consists of two subsystems, the vehicle detection and recognition subsystem and traffic sign detection and recognition subsystem. Each consists of four models: object detection model, object recognition model, object information model, and object tracking model. To detect potential objects on the road, several features are investigated, which include symmetrical shape and aspect ratio of a vehicle and color and shape information of signs. A two-layer neural net is trained to recognize different types of vehicles and a parameterized sign model is established for recognizing a sign. Tracking is accomplished by combining single image frame analysis with consecutive image frame analysis. Single image frame analysis is performed every ten full-size images. The information model obtains the information related to the object, such as time to collision for the object vehicle and relative distance from the traffic signs. Experimental results demonstrate a robust and accurate system in real time object detection and recognition over thousands of frames.
A vision-based object detection and recognition system for intelligent vehicles
1998
12 Seiten, 14 Quellen
Conference paper
English
Vision-based object detection and recognition system for intelligent vehicles [3525-36]
British Library Conference Proceedings | 1998
|Vision-Based Stop Sign Detection and Recognition System for Intelligent Vehicles
Transportation Research Record | 2001
|Vision-Based Stop Sign Detection and Recognition System for Intelligent Vehicles
Online Contents | 2001
|Vision-Based Stop Sign Detection and Recognition System for Intelligent Vehicles
British Library Conference Proceedings | 2001
|