Change Detection and Image Time-Series Analysis 1

7Graph of Characteristic Points for Texture Tracking: Application to Change Detection and Glacier Flow Measurement from SAR Images

Minh-Tan PHAM¹ and Grégoire MERCIER²

¹IRISA Laboratory, Université Bretagne Sud, Lorient, France

²eXo maKina, Digital Technologies, Paris, France

7.1. Introduction

Thanks to the capacity of data acquisition under any atmospheric or weather conditions, one of the most significant applications of synthetic aperture radar (SAR) imagery is to exploit multitemporal data for change detection, which serves for understanding and evaluating land-cover changes occurring after a natural or anthropic disaster, or for identifying and monitoring land-use development over time within certain agricultural, forestry and urban areas (Bovolo and Bruzzone 2007; Del Frate et al. 2008; Mercier et al. 2010; Ban and Yousif 2012). This chapter is dedicated to the context of unsupervised change detection using bitemporal SAR images. To tackle this task, we propose to perform texture tracking based on the characteristic points extracted from the images and modeled by a graph structure. Then, another application using bitemporal SAR data to detect and measure the glacier flows is also investigated based on the proposed texture tracking approach. We first review some related works in unsupervised SAR image change detection in the literature.

Many methods have been proposed to tackle the problem of image change detection. Two systematic surveys on change detection techniques ...