The Multi-Focus-Image-Fusion Method Based on Convolutional Neural Network and Sparse Representation

Abstract

Multi-focus-image-fusion is a crucial embranchment of image processing. Many methods have been developed from different perspectives to solve this problem. Among them, the sparse representation (SR)-based and convolutional neural network (CNN)-based fusion methods have been widely used. Fusing the source image patches, the SR-based model is essentially a local method with a nonlinear fusion rule. On the other hand, the direct mapping between the source images follows the decision map which is learned via CNN. The fusion is a global one with a linear fusion rule. Combining the advantages of the above two methods, a novel fusion method that applies CNN to assist SR is proposed for the purpose of gaining a fused image with more precise and abundant information. In the proposed method, source image patches were fused based on SR and the new weight obtained by CNN. Experimental results demonstrate that the proposed method clearly outperforms existing state-of-the-art methods in addition to SR and CNN in terms of both visual perception and objective evaluation metrics, and the computational complexity is greatly reduced. Experimental results demonstrate that the proposed method not only clearly outperforms the SR and CNN methods in terms of visual perception and objective evaluation indicators, but is also significantly better than other state-of-the-art methods since our computational complexity is greatly reduced.

Raspberry Pi-Embedded Intelligent Control System for Irrigation and Fertilization Based on deep learning

“…In general, the deep learning model has a large number parameters in the deep layer, while some parameters of the deep model have little impact on the performance of the model, so some parameters and convolution layers of the model are redundant, this result is consistent with the conclusion drawn by Wei Yue [11]. For different data sets, different model compression strategies can be used to eliminate unnecessary parameters in the model, and lightweight deployment of deep learning models can be realized in embedded systems.…”

A Survey of Multi-Focus Image Fusion Methods

“…Laplacian pyramid (LP) [9]; gradient pyramid (GD) [10]; contrast pyramid (CP) [11]; region mosaicking on Laplacian pyramids (RMLP) [12]; fast discrete curvelet transform (FDCT) [5]; dual-tree complex wavelet transform (DT-CWT) [13]; nonsubsampled contourlet transform (NSCT) [14]; nonsubsampled shearlet transform (NSST) [15]; cross sparse representation (CSP) [16]; independent component analysis (ICA) [17]; discrete cosine transform (DCT) [18] Boundary segmentation adaptive region-segmentation (ARS) [19]; morphology-based focus measure (MBFM) [20]; content adaptive blurring (CAB) [21]; robust principal component analysis (RPCA) [22]; Markov random field (MRF) [23]; optimal defocus estimation (ODE) [24] Deep learning convolutional neural networks (CNN) [25]; CNN based [26]; PCNN-Pulse Coupled Neural Network (PCNN) [27]; PCNN based [28]; Generative Adversarial Networks (GAN) [29] Combination CTD+SR [30]; NSCT+SR [31]; SF-PAPCNN+NSST+ ISML [32]; CNN+SR [33] 2…”

Gaussian of Differences: A Simple and Efficient General Image Fusion Method

“…HDR images retain details in both the dark and bright regions, which enhances image quality, increases visual fidelity, and improves image analysis in computer vision tasks [ 13 , 14 ]. Multi-focus image fusion is employed to merge multiple images exhibiting distinct focus levels into a singular composite image [ 15 , 16 , 17 , 18 , 19 ]. This results in improved overall sharpness, enhanced depth of field, and enhanced visual perception [ 20 ].…”