A Novel Technique Based on Visual Words Fusion Analysis of Sparse Features for Effective Content-Based Image Retrieval

Table 3

Performance analysis of the proposed technique based on visual words fusion on the Corel-B image collection which is reported using dictionary size of 1000 visual words and features percentage of 50% per image (bold values indicate the best performance).

Performance measures	Proposed technique based on visual words fusion	GMM + mSpatiogram [45]	SQ + spatiogram [3]
MAP	85.20	74.10	63.95
Average recall	17.00	13.80	12.79