Distinctive Order Based Self-Similarity descriptor for multi-sensor remote sensing image matching Amin Sedaghat ⇑ , Hamid Ebadi Faculty of Geodesy and Geomatics Engineering, K.N. Toosi University of Technology, Tehran 19697-64499, Iran article info Article history: Received 15 October 2014 Received in revised form 12 February 2015 Accepted 4 June 2015 Keywords: Image registration Image matching UR-SIFT Local self-similarity Distinctive Order Based Self-Similarity abstract Robust, well-distributed and accurate feature matching in multi-sensor remote sensing image is a diffi- cult task duo to significant geometric and illumination differences. In this paper, a robust and effective image matching approach is presented for multi-sensor remote sensing images. The proposed approach consists of three main steps. In the first step, UR-SIFT (Uniform robust scale invariant feature transform) algorithm is applied for uniform and dense local feature extraction. In the second step, a novel descriptor namely Distinctive Order Based Self Similarity descriptor, DOBSS descriptor, is computed for each extracted feature. Finally, a cross matching process followed by a consistency check in the projective transforma- tion model is performed for feature correspondence and mismatch elimination. The proposed method was successfully applied for matching various multi-sensor satellite images as: ETM+, SPOT 4, SPOT 5, ASTER, IRS, SPOT 6, QuickBird, GeoEye and Worldview sensors, and the results demonstrate its robustness and capability compared to common image matching techniques such as SIFT, PIIFD, GLOH, LIOP and LSS. Ó 2015 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS). Published by Elsevier B.V. All rights reserved. 1. Introduction Image matching is the process of finding corresponding points in two or more images of the same scene and is a crucial process to a wide range of applications such as image registration (Gianinetto, 2012; Parmehr et al., 2014), change detection (Qin and Gruen, 2014; Sadeghi et al., 2013), and to the 3D modeling and mapping sciences (Ahmadi et al., 2010; Ekhtari et al., 2009; Lerma et al., 2013; Mohammadi and Malek, 2014; Soheilian et al., 2013). Local invariant image features have recently received more attention in the field of photogrammetry and remote sensing. Nowadays, most image matching algorithms are based on the local invariant features because of their robustness to geometric and illumination differences. The most of local feature matching algorithm consist of three main steps:  Feature detection, which selects salient features in two images (reference image and input image), such as corners, blobs and regions.  Feature description, which generates feature attributes (‘‘de- scriptors’’ to characterize and match them) using various image properties such as intensity, color, texture, and edge.  Feature matching, which establishes the correspondence between the features in the two images using particular simi- larity measures of their descriptors and then use a consistency check process to remove probable mismatches. The main objective of this article is to develop a local feature matching method for various remote sensing images having differ- ent sensors, resolutions, illuminations and acquisition times. In previous researches, many image matching methods based on local features have been proposed in the literature (Barandiaran et al., 2013; Goshtasby, 2012; Gruen, 2012; Tuytelaars and Mikolajczyk, 2008). The most popular local feature detector and descriptor algo- rithm is the scale invariant feature transform (SIFT) (Lowe, 2004) that uses the DoG (Difference of Gaussian) scale space function and distribution of gradients for detection and description respec- tively. Recently SIFT based method has been widely applied in remote sensing image matching and registration (Joglekar et al., 2014; Sedaghat et al., 2011; Yu et al., 2013). Wang et al. (2012) proposed a robust multisource image auto- matic registration system (MIARS) based on the SIFT descriptor, which uses image division and histogram equalization as pre-processing steps. Han et al. (2012) proposed an automatic http://dx.doi.org/10.1016/j.isprsjprs.2015.06.003 0924-2716/Ó 2015 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS). Published by Elsevier B.V. All rights reserved. ⇑ Corresponding author. E-mail addresses: am.sedaghat@gmail.com (A. Sedaghat), ebadi@kntu.ac.ir (H. Ebadi). ISPRS Journal of Photogrammetry and Remote Sensing 108 (2015) 62–71 Contents lists available at ScienceDirect ISPRS Journal of Photogrammetry and Remote Sensing journal homepage: www.elsevier.com/locate/isprsjprs