A New Statistical Technique for Interpolation of Landsat Images

In this paper, we introduce a new adaptive scheme for interpolation in remote sensing images. In this scheme, we apply an unconditional estimation without any assumption regarding image generality or default weights. Adaption in here is regarding two issues, first, location-dependent interpolation in image and second, adaptive estimation of all pixel weights at each region of image. Proposed scheme by use of this adaptive approach estimates target pixels better than classical and non-adaptive interpolations in practice, such as algorithms implemented in ENVI.

Keywords:

Landsat8 , Markov Random Field (MRF) , Interpolation Algorithms , Adaptive Estimation , Linear Minimum Mean Square Error-estimation (LMMSE)