SATELLITE IMAGE BASED DETECTION OF MISSILE SITES

Abstract

There is an abundance of information in satellite images which is very useful in the military decision making process on the battlefield. It goes without saying that missiles area strategic asset of any nation and likewise their location, structure etc based information is of vital importance. Satellite image processing for valid information extraction is a manually tedious job, therefore there is scope for certain degree of automation as part of the information extraction process from satellite images. A fundamental step in the process of information extraction is that of object detection. Another important element in the information processing workflow is that of Change Detection wherein the point of change detection amongst a time ordered series of images is of interest to the image analyst. This step may be considered as a pre requisite prior to application of the object detection model to images about the change point. The topic is of great interest and has a promising future with a wide variety of applications. This work takes a time ordered set of satellite images and initially detects missile sites as part of the object detection phase followed by locating the point of change as part of the change detection phase. It is hoped that this work will invoke greater interest and initiatives for all researchers in the area of satellite image processing for useful information extraction. Information extraction for an analyst can typically be object detection and change detection. Object detection would focus on classification and identification of targets as early as possible whereas change detection would mean to locate the point of change amongst the time ordered series of images. The scope and application of the tool of Artificial Intelligence is also covered as part of the work.