Journal of Advanced Defense Science & Technology

Journal of Advanced Defense Science & Technology

Camouflage Assessment in Remote Sensing Images Based on Template Matching and UR-SIFT Algorithm

Authors
Abstract
In this paper an efficient method for camouflage assessment in remote sensing images is presented. The proposed method consists of two main parts including template matching algorithm and UR-SIFT (Uniform Robust Scale Invariant Feature Transform) feature extraction algorithm. In the first part, the distinctiveness between target and background based on template matching algorithm is computed and using a Gaussian function a quantitative measure is presented for description of target camouflage. In the second part, the detectability rate of the target is investigated based on UR-SIFT algorithm which is an improved version of well-known SIFT algorithm that has been developed for feature extraction in remote sensing images. If the target would not be extracted using UR-SIFT algorithm then, it can be concluded that target are concealed in the background. Four satellite images included a specific object in four camouflage status are considered to assessment of proposed algorithm. Experimental result indicates the good performance of proposed method in camouflage assessment in remote sensing images.
Keywords

Volume 2, Issue 4 - Serial Number 6
Spring 2011
Pages 295-306

  • Receive Date 30 January 2019
  • Publish Date 22 December 2011