A new regions-of-interest based image retrieval using DWT
Wang Xiangyang, Yang Hongying, Hu Fengli · 2006
In order to overcome the deficiencies of most regions-of-interest (ROI) based image retrieval (such as ROI extracting with not considering the human visual characteristic, not utilizing the location feature of ROI, unreasonable similarity computational model), a new ROI based image retrieval using DWT is proposed. Firstly, the ROI are extracted in DWT domain by using the human visual characteristic and K-mean clustering. Secondly, the local energy of wavelet coefficients of ROI is used for the texture feature, the mean values and standard deviations are used for the color feature, and the weight center coordinate of ROI is used for the location feature. Finally, the average similarity between images is computed according to the ROF above features. Experimental results show that our image retrieval is more accurate and efficient in retrieving the user-interested images when there are ROI in the image (especially for the image with simple background).