Design and Development of Decorrelation Stretch Technique for Enhancing the Quality of Satellite Images with Improved MSE and UIQI in Comparison with Wiener Filter

Abstract

Image enhancement is the process of increasing the quality of an image by enhancing certain features of an image and the resultant image is more suitable than the input image. This research is aimed at developing more efficient image enhancement techniques in the spatial domain. Materials and methods: Image enhancement using decorrelation stretch technique (N=30) is proposed in this work, and this method is compared with Wiener filter (N=30). The sample size used for both groups is 30. Results: The quality enhanced image is measured using Mean Square Error (MSE) and Universal Image Quality Index measure (UIQI). High UIQI and low MSE indicates the better enhanced image. The parameters of MSE (db) and UIQI are having higher values than the parameters of Wiener filter. The mean MSE value of decorrelation is 4.2715 db. Mean MSE of wiener filter is 32.0476 db. The mean UIQI value of decorrelation is 0.5640 db and that of Wiener filter is 0.5766 db. Conclusion: Image enhancement using decorrelation stretch technique is significantly performed better than Wiener filtering based on the experimental results and independent sample T test results using IBM-SPSS software.