Texture feature representation in dynamic environments

This paper presents a novel approach to detect and formulate time varying changes in texture content on a sequence of images registered from a scene by using Gabor wavelets representation. The changes might occur under variable illumination or different location of viewpoint for image registration. It is necessary that such changes might exactly be measured by a vision system. The context of situations such as variable lighting conditions and continuous object movements in dynamic environments can be appropriately interpreted through successful measurement. Natural texture images and synthetic texture images were used for several experiments in this work. These experimental results validated that this approach was effective when there exist orientation and periodic elementary patterns in texture contents.

• Type: Journal
• Year: 2010
• Where: IEEEHPCS
• Authors: Kyeong Deok Woo, Sung Gook Kim, Sung Wook Baik
• Keywords: Applied Computing, IEEEHPCS 2010, Images, Texture Contents, Texture Images