Improved color texture descriptors for remote sensing image retrieval

Abstract

Abstract. Texture features are widely used in image retrieval literature. However, conventional texture features are extracted from grayscale images without taking color information into consideration. We present two improved texture descriptors, named color Gabor wavelet texture (CGWT) and color Gabor opponent texture (CGOT), respectively, for the purpose of remote sensing image retrieval. The former consists of unichrome features computed from color channels independently and opponent features computed across different color channels at different scales, while the latter consists of Gabor texture features and opponent features mentioned above. The two representations incorporate discriminative information among color bands, thus describing well the remote sensing images that have multiple objects. Experimental results demonstrate that CGWT yields better performance compared to other state-of-the-art texture features, and CGOT not only improves the retrieval results of some image classes that have unsatisfactory performance using CGWT representation, but also increases the average precision of all queried images further. In addition, a similarity measure function for proposed representation CGOT has been defined to give a convincing evaluation. © The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI.

Towards Adaptive High-Resolution Images Retrieval Schemes

“…it contains the following classes: agricultural, airplane, baseball diamond, beach, buildings, chaparral, dense residential, forest, freeway, golf course, harbor, intersection, medium residential, mobile home park, overpass, parking lot, river, runway, sparse residential, storage tanks, and tennis court. Studying references (Eptoula, 2014) ) (Shao and al., 2014) (Yang and Newsman, 2012) (Bouteldja and Kourgli, 2015), one can observe that false images are retrieved when different categories share some common textures or/and structures such as those representing buildings, intersection, storage tanks, overpass and tennis court categories. This could be explained by the fact that these categories are more complex containing different structures with different shapes and textures.…”

“…Let one keeps in mind that for each channel, SIFT operator, computed on a vector of 128 elements (Lowe, 1999), needs the buildings of a bag of visual word, whereas CT-DWT (Kingsbury, 2000) and Steerable Pyramid produce a huge vector whose elements (histograms or statistical moments) are derived from the sub images in different orientations at different levels. Also, CGOT vector is constituted of 80 Gabor texture features which give a long vector (Shao and al., 2014). While, we tested the proposed scheme with two features histogram of local variance computed using 16 bins and a modified uniform LBP that produce an histogram on 10 bins constituting after concatenation a feature vector whose length is 26.…”

“…So, their efficiency depends on the choice of the set of visual features and on the choice of the similarity metric that models user perception of similarity. Recent studies (Eptoula, 2014) ) (Shao and al., 2014) (Yang and Newsman, 2012) (Bouteldja and Kourgli, 2015) using a common dataset showed that multi scale feature are more adapted for the retrieval of HRSI.…”

“…Many approaches have been proposed to retrieve low and mid-satellite images using their content such as region level semantic features mining (Lu and al., 2012), Knowledge-driven information mining (KIM) (Daschiel and al., 2003) , texture model (Aksoy and al., 2013) entropy-balanced bitmap (EBB) tree (Scott and al., 2011). High resolution satellite retrieval schemes use different features according to colour (spectral) features (Bag and Guo, 2004), texture features (Yang and Newsman, 2012) ) (Shao and al., 2014) and structure features (Yang and Newsman, 2012). Most of these approaches are expressed by visual examples in order to retrieve from the database all the HRSI that are similar to the examples and achieved a satisfactory success for some types of categories.…”

Towards Adaptive High-Resolution Images Retrieval Schemes

“…Many approaches have been proposed to retrieve low and mid-satellite images using their content such as region level semantic features mining (Lu and al., 2012), Knowledge-driven information mining (KIM) (Daschiel and al., 2003) , texture model (Aksoy and al., 2013) entropy-balanced bitmap (EBB) tree (Scott and al., 2011). High resolution satellite retrieval schemes use different features according to colour (spectral) features (Bag and Guo, 2004), texture features (Yang and Newsman, 2012) ) (Shao and al., 2014) and structure features (Yang and Newsman, 2012). Most of these approaches are expressed by visual examples in order to retrieve from the database all the HRSI that are similar to the examples and achieved a satisfactory success for some types of categories.…”