DROUGHT PREDICTION AND ANALYSIS OF WATER LEVEL BASED ON SATELLITE IMAGES USING DEEP CONVOLUTIONAL NEURAL NETWORK

Abstract

Resolution enhancement (RE) schemes (which are not based on wavelets) suffer from the Disadvantage of losing high-frequency contents (which results in blurring). The discrete-wavelet-transform-based (DWT) RE scheme generates artifacts (due to a DWT shift-variant property). A wavelet-domain approach based on dual-tree complex wavelet transform (DT-CWT) and nonlocal means (NLM) is proposed for RE of the satellite images. A satellite input image is decomposed by DT-CWT (which is nearly shift invariant) to obtain high-frequency sub- bands. The highfrequency and the low-resolution (LR) input image are interpolated using the Lanczos interpolator. The highfrequency sub-bands are passed through an NLM Filter to cater for the artifacts generated by DT-CWT (despite of it’s nearly shift invariance). The F i ltered high-frequency sub-bands and the LR input image are combined using inverse DT-CWT to obtain a resolution-enhanced image. Objective and subjective analyses reveal superiority of the proposed technique over the conventional and state-of-the-art RE techniques.