Text plates detection and recognition techniques used for an autonomous robot navigating in indoor environments

Abstract

The vision system of our autonomous robot possessing the text plate detection and recognition abilities is the main equipment to capture surrounding information. By virtue of a PTZ camera, we can obtain the sequential images of indoor environments. As the autonomous robot moving forward, the obtained sequential images are fed to execute text plate detection. We analyze the color information in these images to determine candidate regions, and observe the edge points in each candidate region after edge detection. Some regions will be filtered out if they have too few edge points. We will recognize the text plate by means of the linear discriminant analysis method. When the recognition result matches with the predefined target, the autonomous robot will carry out the corresponding actions. In the experiments, for uniform contrast text plates, the average text plate detection rate is more than 90%, and the average text plate recognition rate is more than 91% in two different office environments. The robot vision system works by 2 frames per second on an average, and there contains 640×480 pixels per frame captured by our camera.