Interactive differential evolution for user-oriented image retrieval system

Abstract

Large amounts of image data have been produced on the Internet over the past several years. As a kind of effective retrieval way, the content-based image retrieval (CBIR) has attracted more and more attention. To improve the preciseness, most CBIR systems emphasize on finding the best representation for different image features. However, the semantic gap between visual description and user expectations is hard to handle. The relevance feedback technique can use relevance information to alleviate this problem. This paper describes a CBIR framework based on interactive differential evolution which uses a technique of combing the global and the local retrieval strategy to help users retrieve their preferred images in a user-oriented way. Experimental results show that the proposed framework increases the accuracy, and it outperforms the recent frameworks based on GAs.

Pattern retrieval of yarn‐dyed plaid fabric based on modified interactive genetic algorithm

“…The visual features proposed in this article were compared with the low‐level features used in previous studies. The color features involve the mean value and the SD of pixel colors, 20 image binary bitmap, 21 and the combination of them 19 . The texture features were all extracted by gray level co‐occurrence matrix.…”

“…The color feature was extracted by HS color histogram and the texture feature was extracted by granulometry cumulative distributions. In other studies, the mean value and the SD of pixel colors 19,20 and image binary bitmap 19,21 were extracted as color features and the texture features were extracted by gray level co-occurrence matrix [19][20][21] for IGA-based image retrieval. The pattern retrieval of existing yarn-dyed plaid fabric has not been reported yet.…”

An Interactive Differential Evolution Algorithm Based on Backtracking Strategy Applied in Interior Layout Design

“…The iterative process of the algorithm is shown in Figure 2. In recent years, interactive differential evolution algorithm has been applied in image retrieval [27], image enhancement [28], image filter [29] and other aspects.…”