Adaptive chebyshev fusion of vegetation imagery based on SVM classifier

Abstract

A novel adaptive image fusion method by using Chebyshev polynomial analysis (CPA), for applications in vegetation satellite imagery, is introduced in this paper. Fusion is a technique that enables the merging of two satellite cameras: panchromatic and multi-spectral, to produce higher quality satellite images to address agricurtural and vegetation issues such as soiling, floods and crop harvesting. Recent studies show Chebyshev polynomials to be effective in image fusion mainly in medium to high noise conditions, as per real-life satellite conditions. However, its application was limited to heuristics. In this research, we have proposed a way to adaptively select the optimal CPA parameters according to user specifications. Support vector machines (SVM) is used as a classifying tool to estimate the noise parameters, from which the appropriate CPA degree is utilised to perform image fusion according to a look-up table. Performance evaluation affirms the approach’s ability in reducing the computational complexity to perform fusion. Overall, adaptive CPA fusion is able to optimize an image fusion system’s resources and processing time. It therefore may be suitably incorporated onto real hardware for use on vegetation satellite imagery.