Image feature extraction for image database content description: Indoor room images

Abstract

This paper presents the theoretical concept and design of image feature extraction by using selected hybrid image feature extraction techniques; the combination of Scale-Invariant Feature Transform (SIFT), Elastic Bucket Tries and Fuzzy-Rule Based techniques. We show the advantages and disadvantages of the combination of those techniques, tested for indoor room images. The proposed indoor images style was Malay, Chinese and Indian. The result is hoped to contribute a significant idea to the development of technology in classifying the indoor room images based on other styles such as Bali, English etc.