A TV logo detection and recognition method based on SURF feature and bag-of-words model
Jingmeng He, Xie Yuxiang, Luan Xidao, Niu Xiao, Xin Zhang · 2016
TV logo is one of the important symbols of TV station. Nowadays TV logo recognition has been widely used in various fields, such as video analysis, video surveillance, video management and so on. According to TV logo's characteristics, a TV logo detection and recognition algorithm based on SURF feature and bag-of-visual-words model is proposed in this paper. First, obtain binary logo images by using the automatic image segmentation method based on Otsu in the region of interest, and then execute AND operation with original images, thereby completing the logo extraction. Second, extract logo images' SURF features, and then use k-means clustering algorithm to construct a visual dictionary, and ultimately use support vector machine (SVM) to complete logo recognition. Experimental results show that the proposed method has high recognition accuracy, strong anti-interference ability and good practicality.