Identification of Change Detection for Satellite Images using Ant Colony Optimization

Supervision of Earth surface with the help of satellite images helps in land use land cover, resource planning & management, change detection and agricultural development areas. Satellites images are great repository of information and extraction of information from these unclear images is a challenging task. In order to identify the objects and boundaries in satellite images, various digital image based processing techniques like Sobel, Canny and Prewitt have been explored in this paper. Further, Ant Colony Optimization has been used to obtain the optimized images. Hybrid techniques viz.- AC0-Sobel, ACO-Canny and ACO-Prewitt are implemented and the results are quantitatively validated with the help of entropy and PIQE values of output optimized images. The results indicates that ACO-Canny hybrid technique yields better quality satellite images as compared to ACO-Sobel, ACO-Prewitt, Sobel, Canny and Prewitt edge detection techniques.