A Novel Evolutionary Approach for Optimizing Content-Based Image Indexing Algorithms

Abstract

Optimization of content-based image indexing and retrieval (CBIR) algorithms is a complicated and time-consuming task since each time a parameter of the indexing algorithm is changed, all images in the database should be indexed again. In this paper, a novel evolutionary method called evolutionary group algorithm (EGA) is proposed for complicated time-consuming optimization problems such as finding optimal parameters of content-based image indexing algorithms. In the new evolutionary algorithm, the image database is partitioned into several smaller subsets, and each subset is used by an updating process as training patterns for each chromosome during evolution. This is in contrast to genetic algorithms that use the whole database as training patterns for evolution. Additionally, for each chromosome, a parameter called age is defined that implies the progress of the updating process. Similarly, the genes of the proposed chromosomes are divided into two categories: evolutionary genes that participate to evolution and history genes that save previous states of the updating process. Furthermore, a new fitness function is defined which evaluates the fitness of the chromosomes of the current population with different ages in each generation. We used EGA to optimize the quantization thresholds of the wavelet-correlogram algorithm for CBIR. The optimal quantization thresholds computed by EGA improved significantly all the evaluation measures including average precision, average weighted precision, average recall, and average rank for the wavelet-correlogram method.

Á trous gradient structure descriptor for content based image retrieval

“…Due to variety in contents these databases are being used by researchers in various scientific articles of CBIR. Various performance measures like precision, recall, and rank, etc., are calculated to compare the performance of proposed methods with some of already published papers [13,14,16] (see Appendix). Precision is defined in terms of number of relevant images retrieved out of total number of retrieved images considered.…”

“…The authors have first applied PM1 for CBIR application and analyzed its performance with respect to published literature. It is observed that PM1 results are better than [13,14] and [16]. Tables 1, 2, 4 and 5 shows parameter P wt , P wt_std , P avg , P std , R avg , R std , C avg and C std values for OQWC [14], Subrahmanyam et al [16], PM1 and PM2, respectively.…”

“…Saadatmand et al [14] has improved the performance of WC by optimizing the wavelet coefficients. Gonde et al [15] proposed the texton co-occurrence matrix for image retrieval.…”

Content- based Image Retrieval Approach using Three Features Color, Texture and Shape

“…proposed an issue for effective and efficient content based image retrieval by presenting a novel indexing and retrieval methodology that combines color, shape and texture, information for the indexing and retrieval [8]. Clustering, particularly fuzzy Cmeans (FCM)-based clustering and its variants, have been extensively used for segmentation due to their simplicity and fast convergence [20] [10] and shown that the performance improvement can be obtained by optimizing the quantization thresholds [11] using genetic algorithm for CBIR application. Texture retrieval is a limb of texture analysis that has attracted wide attention from industries since this is well suited for the identification of products such as ceramic tiles, marble, parquet slabs, etc.…”

“…They are statistical and structural approaches [10]. In structural based texture method, the surface pattern is repetitive such as floor design that contains the same pattern [11]. In statistical texture, the surface pattern is not regularly repetitive such as different flower objects in a picture that normally have similar properties but not exactly the same [12].…”