A Differential Evolutionary Multilevel Segmentation of Near Infra-Red Images Using Renyi’s Entropy
In recent years remote sensing image processing has got some intense attention of the researchers for its utility in land cover study, natural calamity detection, object tracking etc. In case of remote sensing image processing, the primal objective is to sub divide the image into more than one segment. In doing so, Multi-level thresholding based image segmentation techniques play an useful role in accomplishing this critical task. Endeavor of this paper is to focus on obtaining the optimal multiple threshold points from a LISS III Near Infra-Red (NIR) band by employing Renyi’s Entropy. Moreover, a state-of-art meta-heuristics like Differential Evolution (DE) is incorporated to acquire optimal threshold values in reduced computational time with precision.
KeywordsMultilevel Image Segmentation Remote Sensing Images Renyi’s Entropy Differential Evolution LISS-III Near Infra-Red (NIR)
Unable to display preview. Download preview PDF.
- 2.Sahoo, P.K., Arora, G.: A thresholding method based on 2-D Renyi’s entropy. Pattern Recognition, 1149–1161 (2004)Google Scholar
- 4.Storn, R., Price, K.V.: Differential Evolution - a simple and efficient adaptive scheme for global optimization over continuous spaces. Technical Report TR-95-012, ICSI (1995), http://http.icsi.berkeley.edu/~storn/litera.html
- 6.Sarkar, S., Patra, G.R., Das, S.: A Differential Evolution Based Approach for Multilevel Image Segmentation Using Minimum Cross Entropy Thresholding. In: Panigrahi, B.K., Suganthan, P.N., Das, S., Satapathy, S.C. (eds.) SEMCCO 2011, Part I. LNCS, vol. 7076, pp. 51–58. Springer, Heidelberg (2011)CrossRefGoogle Scholar
- 7.Rényi, A.: On measures of information and entropy. In: Proceedings of the 4th Berkeley Symposium on Mathematics, Statistics and Probability, pp. 547–561 (1960)Google Scholar