Performance Comparison of various levels of Fusion of Multi-focused Images using Wavelet Transform

Abstract

The fast development of digital image processing leads to the growth of feature extraction of images which leads to the development of Image fusion. Image fusion is defined as the process of combining two or more different images into a new single image retaining important features from each image with extended information content. There are two approaches to image fusion, namely Spatial Fusion and Transform fusion. In Spatial fusion, the pixel values from the source images are directly summed up and taken average to form the pixel of the composite image at that location. Transform fusion uses transform for representing the source images at multi scale. The most common widely used transform for image fusion at multi scale is Wavelet Transform since it minimizes structural distortions. But, wavelet transform suffers from lack of shift invariance & poor directionality and these disadvantages are overcome by Stationary Wavelet Transform and Dual Tree Wavelet Transform. The conventional convolution-based implementation of the discrete wavelet transform has high computational and memory requirements. Lifting Wavelets has been developed to overcome these drawbacks. The Multi-Wavelet Transform of image signals produces a non-redundant image representation, which provides better spatial and spectral localization of image formation than discrete wavelet transform. And there are three levels of image fusion namely Pixel level, Area level and region level. This paper evaluates the performance of all levels of multi focused image fusion of using Discrete Wavelet Transform, Stationary Wavelet Transform, Lifting Wavelet Transform, Multi Wavelet Transform, Dual Tree Discrete Wavelet Transform and Dual Tree Complex Wavelet transform in terms of various performance measures.

Comparison of DCT, SVD and BFOA based multimodal biometric watermarking systems

“…The iris pattern was encoded by Gabor filters after the process of segmentation and normalization. Kannan et al [3] evaluated the performance of all levels of multifocused image fusion using Discrete Wavelet Transform, Stationary Wavelet Transform, Lifting Wavelet Transform, Multi Wavelet Transform, Dual Tree Discrete Wavelet Transform and Dual Tree Complex Wavelet transform in terms of various performance measures. Shah et al [4] presented a novel fusion rule which can efficiently fuse multifocus images in the wavelet domain by taking a weighted average of the pixels.…”

Image Fusion Techniques: A Review

“…Stationary wavelet transform (SWT) [1] is modified DWT where this drawback is removed by removing the down samplers and up samplers in DWT and up sampling the filter coefficients by a factor of 2(j-1)in the jth level of the algorithm. Borwonwatanadelok et al [32] , Somkait et al [33] and kannan et al [34] have shown that This method provides good result at level 2 of decomposition but it is very time consuming.…”

“…Stationary Wavelet Transform(SWT) [32][33][34] This method provides good result at level2 of decomposition It is time consuming…”

Improved Block Based Feature Level Image Fusion Technique Using Contourlet with Neural Network

“…A. Toet proposed an algorithm for image fusion by a ratio of low pass pyramid [5]. K. Kannan et al evaluated the performance of all the levels of multi-focused images using different wavelet transforms [6]. Dong et al discussed various advances in multi-sensor data fusion [7].…”