Satellite Image Segmentation using Swarm Intelligence

The most complex and dense images are the satellite and remote sensing images. In this era, numerous nature-based optimization algorithms are proposed and analyzed for achieving multi level thresholding using different entropy functions. Two algorithms MABC and CS algorithms have been proved their strength to achieve global optimum threshold using different objective functions especially Kapur's entropy. The drawback, which has both the algorithm, is their complexity which is higher than other algorithms available in literature. In this paper, both the algorithms have been discussed in reference of segmentation of satellite and remote sensing images. All the pros and cons of these two algorithms are also taken into consideration to analyse the performance. The future scope is also discussed to propose a new approach which may be more efficient than these two. Using the analysis of these two algorithms we propose a methodology in which we enhance the segmented image quality and computation of optimum thresholding. This will provide some important conclusions regarding approach.