A Bk Subproduct Approach for Scene Classification

Scene classification has received much attention in the recent past. How images can be efficiently classified becomes an important research topic. Most of the stateof-the-art solutions assumed that scene classes are mutually exclusive. But, different human beings have different perceptions, and different people tend to respond inconsistently even to the same scene images. Hence, a scene image may belong to multiple classes. In this paper, we propose a BK subproduct approach to address this issue. The advantages are two-folds. Firstly, our approach is able to model the non-mutually exclusive data. Secondly, our classification result is not binary. Instead, it classifies each scene image as a combination of different classes using membership function. Empirical results using a standard dataset had shown the effectiveness and robustness of the proposed method.