Vsevolod Yugov - Tokyo Institute Of Technology, Japan, Dept. of Information Processing
Itsuo Kumazawa - Dr. of Engineering, Imaging Science and Engineering Laboratory Tokyo Institute of Technology, Japan

Feature point matching and texture recognition are two of the most important problems in the image processing. Recently, several new approaches to these problems using simple localfeatures and semi-naive Bayesian classification scheme have been developed. In our paper, we show how to enhance thesetechniques further by combining them with Support Vector Machines using online learning techniques. The resultingalgorithm is simple, robust and can be adapted to various tasks inimage processing. Furthermore, we demonstrate the advantages of our method by using it to achieve real-time texture recognition on a mobile device by utilizing parallel processing capabilities afforded by the device GPU

Image processing, Mobile device, GPU, Texture recognition, FERNS