Study of image retrieval and classification based on adaptive features using genetic algorithm feature selection Chuen-Horng Lin a,⇑ , Huan-Yu Chen b , Yu-Shung Wu a a Department of Computer Science and Information Engineering, National Taichung University of Science and Technology, No. 129, Sec. 3, Sanmin Rd., Taichung, Taiwan, ROC b Department of Electrical Engineering, National Chung Hsing University, No. 250 Kuo Kuang Rd., Taichung, Taiwan, ROC article info Article history: Available online 6 May 2014 Keywords: Color feature Texture features Genetic algorithms Feature selection Support vector machine abstract This paper proposes a genetic algorithm feature selection (GAFS) for image retrieval systems and image classification. Two texture features of adaptive motifs co-occurrence matrix (AMCOM) and gradient histo- gram for adaptive motifs (GHAM) and color feature of an adaptive color histogram for K-means (ACH) were used in this paper. In this paper, the feature selections have adopted sequential forward selection (SFS), sequential backward selection (SBS), and genetic algorithms feature selection (GAFS). Image retrieval and classification performance mainly build from three features: ACH, AMCOM and GHAM, where the classi- fication system is used for two-class SVM classification. In the experimental results, we can find that all the methods regarding feature extraction mentioned in this study can contribute to better results with regard to image retrieval and image classification. The GAFS can provide a more robust solution at the expense of increased computational effort. By applying GAFS to image retrieval systems, not only could the number of features be effectively reduced, but higher image retrieval accuracy is elicited. Ó 2014 Elsevier Ltd. All rights reserved. 1. Introduction With the rapid development of information technology and multimedia technology, many electronic imaging equipments have become popular, resulting in a growing amount of multimedia information. Various image collections and databases have led to the rapid growth of the image retrieval field, a challenging and expanding research area. Thus, determining how to effectively and efficiently retrieve a desired image from a constantly growing image database has become an important issue. Traditional image retrieval systems are based on the features of the original data (Gong, Zhang, Chuant, & Skauuchi, 1994; Gu, Panda, & Haque, 2001), such as: file name, note title, keyword and indexing icon. When applied to large scale image databases, these features become troublesome, time-consuming and even inadequate in regard to describing image contents. To improve the accuracy of the image retrieval systems, it is important to have a proper image feature set that describes the precise contents of an image. The more suitable the image features that are set, the higher the retrieval accuracy that results! Similarity computation step models employ image similarity based on combinations of various features extracted from images. In recent years, image retrieval systems have been based on image contents which are most com- monly used, i.e. color, texture, spatial relationships, shape and oth- ers. The focus has also been narrowed in developing new techniques, wherein effective retrieval and browsing of large digi- tal image libraries is based on automatically derived imagery features. Many feature-based image retrieval systems have been pro- posed (Swain & Ballard, 1991; Rui & Huang, 1999; Brnuelli & Mich, 2001; Chun, Seo, & Kim, 2003; Ko & Byun, 2005; Hurtut, Gousseau, & Schmitt, 2008; Lin, Chen, & Chan, 2009; Lin & Lin, 2010; Haralick, Shanmugam, & Dinstein, 1973; Huang & Dai, 2003; Jhanwar, Chaudhurib, Seetharamanc, & Zavidovique, 2004; Moghaddam, Khajoie, Rouhi, & Tarzjan, 2005; Hafiane & Zavidovique, 2008; Liu & Yang, 2008; Wei, Li, Chau, & Li, 2009). For instance, colors (Swain & Ballard, 1991; Rui & Huang, 1999; Brnuelli & Mich, 2001; Chun et al., 2003; Ko & Byun, 2005; Hurtut et al., 2008; Lin et al., 2009; Lin & Lin, 2010), textures (Lin et al., 2009; Lin & Lin, 2010; Haralick et al., 1973; Huang & Dai, 2003; Jhanwar et al., 2004; Moghaddam et al., 2005; Hafiane & Zavidovique, 2008; Liu & Yang, 2008), spatial relations (Hurtut et al., 2008; Lin, Chan, Chen, Huang, & Chang, 2011) and shapes (Wei et al., 2009) have been extensively applied to the task of image retrieval but the results have garnered limited effects on dis- crimination. Color attribute analysis (Swain & Ballard, 1991; Rui & Huang, 1999; Brnuelli & Mich, 2001; Chun et al., 2003; Ko & Byun, http://dx.doi.org/10.1016/j.eswa.2014.04.033 0957-4174/Ó 2014 Elsevier Ltd. All rights reserved. ⇑ Corresponding author. Tel.: +1 886 4 22195502. E-mail addresses: linch@nutc.edu.tw (C.-H. Lin), d9564104@mail.nchu.edu.tw (H.-Y. Chen), iay128@gmail.com (Y.-S. Wu). Expert Systems with Applications 41 (2014) 6611–6621 Contents lists available at ScienceDirect Expert Systems with Applications journal homepage: www.elsevier.com/locate/eswa