Advanced Change Detection in Satellite Images using DWT
Rekha Jadhav

Rekha Jadhav, Asst. Prof. G. H. Raisoni Institute of Engineering and Technology, Pune, (Maharashtra), India.
Manuscript received on 20 November 2014 | Revised Manuscript received on 30 November 2014 | Manuscript published on 30 November 2014 | PP: 1-5 | Volume-3 Issue-5, November 2014 | Retrieval Number: E1226113514/2014©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: The detection of changes occurring on the earth surface through the use of multitemporal remote sensing images is one of the most important applications of remote sensing technology. In several applications (e.g disaster management, deforestation, crop development, urban growth), the remote sensing-based change detection using multitemporal satellite images is crucial. Particularly, in disaster management cases, the fast and accurate detection of affected regions in multitemporal images acquired at two different time instances, i.e., before and after the disaster, plays a very essential role in taking timely and appropriate decisions. In such cases the change detection method to produce the change results with almost no manual interventions in a reasonable time interval is very important. In this paper, a robust unsupervised change-detection method is proposed. Here, two multitemporal images are taken as input images, geometric and radiometric corrections are performed on one image relative to the other image. Discrete Wavelet Transform (DWT) is applied one both images and then image difference is taken by subtraction. Then features are extracted from difference image using Principal Component Analysis (PCA). Changed and unchanged pixels are separated then by using k-means clustering algorithm with k=2.
Keywords: Remote Sensing, Satellite Images, Change detection, Image differencing, Unsupervised Approach, DWT, PCA, k-means clustering.

Scope of the Article: Clustering