| Summary: | 碩士 === 國立臺灣科技大學 === 資訊工程系 === 97 === In recent years, many researchers do the best of their abilities in the development of service robots. The main purpose of these researches is to assist people deal with routine matters. In our daily life, many supports could be offered from service robots; for example, home security, entertainment, and delivery. In an indoor environment, there is an issue worthy to be investigated that how to guide robots to reach a destination and embed them some intelligence to handle day-and-day works. In this thesis, we would present text plate detection and recognition techniques used for an autonomous robot navigating in indoor environments. We would use text plates to provide the surrounding information and achieve the goal of navigation. Furthermore, some given tasks could be accomplished by the autonomous robot to supply convenient services in an indoor environment such as an office room.
The vision system of our autonomous robot which possesses the text plate detection and recognition abilities is the main equipment to capture surrounding information. By use of a PTZ camera, we could obtain the sequential images of indoor environments. In the experiments, we would adopt the text plates that have high contrast of image intensity to provide the information of region segmentation. As the autonomous robot moving forward, the obtained sequential images are used in 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 would be filtered out because the number of edge points is too few in them. We would recognize the text plate by applying the linear discriminant analysis method. When the recognition result matches with the predefined target, the autonomous robot would execute the corresponding actions.
The experimental results reveal that the text plate detection rate is larger than 93% under the influence of lights and shadows. The detection rate is higher than 90% when the text plates are occluded with blots. Besides, the average text plate recognition rate is about 93.5%, where every letter is counted in statistics. In our system, each captured frame contains 640×480 pixels, and the procedures of text plate detection and recognition work by two frames per second.
|
|---|
