Using adaboost and SIFT for capturing Chinese text content of natural scene image in mobile communication network

The Chinese text in natural scene images which are transmitted in mobile communication network usually contains various and important information. Therefore, capturing Chinese text in these images can analyze the content of these images, and improve the commercial value and security of our network. The paper proposes an efficient and effective method to capture the Chinese text in images which were taken by mobile phone in natural scene. The detection and the location of the target text are achieved by AdaBoost, which can eliminate background noise; the SIFT feature, which is proper to the characteristics of Chinese text in image, is applied to capture the content of the located text. The method has a good practical application value. As the lack of experimental samples, further research is needed to improve.