LDA Feature Selection for Satellite Image Fusion in HAAR Wavelet

Anfal Hazim Abdullah, Dr. E. Sreenivasa Reddy

Abstract: Image fusion is a technique that integrate complimentary details from multiple input images such that the new image give more information and more suitable for the purpose of human visual perception. Major technical constraints like minimum data storage at satellite platform in space, less bandwidth for communication with earth station, etc. limits the satellite sensors from capturing images with high spatial and high spectral resolutions simultaneously. To overcome this limitation, image fusion has proved to be a potential tool in remote sensing applications which integrates the information from combinations of panchromatic, multispectral or hyperspectral images, intended to result in a composite image having both higher spatial and higher spectral resolutions. This paper presents new algorithm to fuse the multi-image in four steps. First discrete wavelet transform (DWT) for time to frequency conversion, feature extraction, feature selection based on LDA and finally classify the feature level fusion. Here LDA will select best feature of the multiple image to fuse the image.

Keywords: Image fusion, Relative spectral contribution methods, Component substitution, Multi-resolution analysis, Quality metrics for performance evaluation, LDA feature selection