Video Images Of This Positioning Of The Chinese Extraction Method

Text is part of the important information in videos and images. Text detrection and recognition in videos and images can help a lot in video contern analysis and understanding, since text can provide concise and direct description of the stories presented in the videos and images. Such as the unique license plate number showed in the Surveillance video can be used for searching the target vehicle if we can get the recognized text. So text detection and analysis is important in video analysis. Thus, how to localize and extract from video and image data under complex-background is one of the hot topics.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. We notice that there is strong color and brightness contrast between the text area and its background, where thers is rich information of corners. So, corners detection is first used in images and video frames, getting the corner map, the we can get the candidate text region by using mathematic morphology. Experimental results show that the algorithm is accurate and spend little time when the text background is simple.Then, this paper discuss the methods of threshold segmentation and realize two classic methods of text segmentation which are based on wavelet transform in combination with support vector machine and wavelet transform in combination with K-Means clustering. Experimental results show that both algorithms are succeed when the background is complicated, even if the text in images has different style, size or position. In the end of this paper, we simply do some research about the theory of characher recognition.