Color Image And Video Text Extraction Algorithm

Text detection and recognition in images and videos can help a lot in video content analysis and understanding, since text can provide concise and direct description of the stories presented in the videos. In digital news videos, the superimposed captions usually present the involved person's name and the summary of the news event. Hence, the recognized text can become a part of index in a video retrieval system.Text extraction in images and videos is a key issue in intelligent information processing. It has received a lot of attention in the academic field of computer vision. A lot of methods have been proposed: based on edge, based on texture, based on region. However, text in images and videos presents a lot of challenges because of low resolution, complex background, lighting variation, and unrestricted pose, shape and color.In this paper, an algorithm for finding text in images and video frames is proposed based on the texture character of text that is called the corners of the text, especially graphic text, are rich of corners. First, corners detection is used in images and video frames, getting the corner map. Second, corner refinement, eliminating the isolate corners and the corners not detected from text. Third, the remained corners in the corner map are merged to form the candidate text region using mathematic morphology. Finally, text region verification is taken to reduce false alarms. Experimental results show that the algorithm is accurate and spend little time when the text background has not many corners.Finally, We also discuss the methods of threshold segmentation and the theory of character recognition in this paper.