Effective Multifocus Image Fusion Based on HVS and BP Neural Network

Abstract

The aim of multifocus image fusion is to fuse the images taken from the same scene with different focuses to obtain a resultant image with all objects in focus. In this paper, a novel multifocus image fusion method based on human visual system (HVS) and back propagation (BP) neural network is presented. Three features which reflect the clarity of a pixel are firstly extracted and used to train a BP neural network to determine which pixel is clearer. The clearer pixels are then used to construct the initial fused image. Thirdly, the focused regions are detected by measuring the similarity between the source images and the initial fused image followed by morphological opening and closing operations. Finally, the final fused image is obtained by a fusion rule for those focused regions. Experimental results show that the proposed method can provide better performance and outperform several existing popular fusion methods in terms of both objective and subjective evaluations.

Dual-Tree Complex Wavelet Transform and Image Block Residual-Based Multi-Focus Image Fusion in Visual Sensor Networks

“…For the high frequency coefficients, the most popular fusion rule is to select the coefficients with larger absolute values, but this rule does not take any consideration of the surrounding pixels. The SML operator is developed to provide local measures of the quality of image focus [ 29 ]. In [ 33 ], it is proved that the SML is very efficient in the transform domain.…”

“…The framework is divided into visual contrast based DTCWT based initial fusion and block residual based final fusion processes. In the visual contrast based DTCWT-based initial fusion process, the Sum-Modified-Laplacian (SML)-based visual contrast [ 29 ] and SML [ 30 ] are employed as the rules for low- and high-frequency coefficients in DTCWT domain, respectively. Using this model, the most important feature information is selected in the fused coefficients.…”

Automated Multimodal image fusion for brain tumor detection

“…Mutual information provides the information quantity detail of input images, which are merged in the resultant image. The highest Mutual information represents the effectiveness of the IF technique [1]. This metric is represented as…”

“…It is designed by modeling any contrast distortion and radiometric. It is combination of the luminance image distortion and combination of contrast distortion, loss correlation and structure distortion between source images and the final image [1,9]. This metric is defined as follow:…”

“…As per literature [15][16][17][18][19][20], many IF frameworks have been developed for the fusion of diverse images in medical domain. It is generally categorized as pixel, feature and decision level [1]. Pixel level image fusion is performed at the lowest level which is the simplest of all fusion methods.…”