Image Parsing with Superpixels and Gray Image Analysis
Yang Feng, Zheng Ma, Mei Hua Xie · 2018
This paper presents an effective system with the aim of solving the problem of image parsing. In order to make full use of information and characteristics in the input image(colored), the gray image is calculated from the color image, and the superpixels of both the gray image and the color one are obtained. After feature extraction, the features abstracted from color image and gray image are fused together to construct the final features. Then, the likelihood ratio scores of the input image and the similar images in the retrieval set are calculated according to these final features and the final labeling result are obtained through MRF analysis. The effectiveness of the proposed system is tested on the public dataset of SIFT Flow and experimental result shows that the proposed method in this paper is efficient.