A Real Time Object Recognition System Based on Feature Matching
基于特征匹配的目标实时识别

A real time object recognition system based on feature maaching is described in this paper. The system can recognize more than 29 object models or more than 11 plane models with similar shapes. The results of recognition are independent of translation, rotation and scale. The experiments show that the correct recognition rate is greater than 98%, and the recognition time is less than 34 ms which meet the need for recognizing the moving objects in real time.