A new bag of words model based on fuzzy membership for image description
Yanshan Li, Weixin Xie, Zhijian Gao, Qinghua Huang, Yujie Cao · 2014
Bag of Words (BoW) as an efficient approach to describing the images has been attracting more and more attention. However, in traditional BoW, the maps between words in codebook and features extracted from images are ambiguous. We propose a new type of BoW based on Gaussian membership function (Gaussian-BoW) to describe images. In Gaussian-BoW, the codebook is obtained by using k-means like the traditional BoW. Then, words are assigned to the feature with Gaussian membership values. At last, histogram is generated by adding up the fuzzy membership values of each word to describe the images. The experimental results show that the proposed Gaussian-BoW outperforms traditional BoW for image description.