Enhancement of Crop Images Using Image Fusion Method in Transform Domain

A new multifocus image fusion method is proposed. Two image blocks are selected by sliding the window from the two source images at the same position, discrete cosine transform (DCT) is implemented, respectively, on these two blocks, and the alternating component (AC) energy of these blocks is then calculated to decide which is the well-focused one. In addition, block matching is used to determine a group of image blocks that are all similar to the well-focused reference block. Finally, all the blocks are returned to their original positions through weighted average. The weight is decided with the AC energy of the well-focused block. Experimental results demonstrate that, unlike other spatial methods, the proposed method effectively avoids block artifacts. The proposed method also significantly improves the objective evaluation results, which are obtained by some transform domain methods.

Download Full-text

Medical Image Fusion Method Based on Coupled Neural P Systems in Nonsubsampled Shearlet Transform Domain

International Journal of Neural Systems ◽

10.1142/s0129065720500501 ◽

2020 ◽

Vol 31 (01) ◽

pp. 2050050 ◽

Cited By ~ 1

Author(s):

Bo Li ◽

Hong Peng ◽

Xiaohui Luo ◽

Jun Wang ◽

Xiaoxiao Song ◽

...

Keyword(s):

Image Fusion ◽

Medical Images ◽

Low Frequency ◽

Transform Domain ◽

Fusion Method ◽

Shearlet Transform ◽

Nonsubsampled Shearlet Transform ◽

Model Combining ◽

Fusion Methods ◽

Image Fusion Method

Coupled neural P (CNP) systems are a recently developed Turing-universal, distributed and parallel computing model, combining the spiking and coupled mechanisms of neurons. This paper focuses on how to apply CNP systems to handle the fusion of multi-modality medical images and proposes a novel image fusion method. Based on two CNP systems with local topology, an image fusion framework in nonsubsampled shearlet transform (NSST) domain is designed, where the two CNP systems are used to control the fusion of low-frequency NSST coefficients. The proposed fusion method is evaluated on 20 pairs of multi-modality medical images and compared with seven previous fusion methods and two deep-learning-based fusion methods. Quantitative and qualitative experimental results demonstrate the advantage of the proposed fusion method in terms of visual quality and fusion performance.

Download Full-text

A Novel Multi-focus Image Fusion Method using Pulse Coupled Neural Network in Nonsubsampled Contourlet Transform Domain

Proceedings of the 2015 International Industrial Informatics and Computer Engineering Conference ◽

10.2991/iiicec-15.2015.60 ◽

2015 ◽

Author(s):

Chaoben Du ◽

Haifeng Yan ◽

Shesheng Gao ◽

Gaoge Hu

Keyword(s):

Neural Network ◽

Image Fusion ◽

Contourlet Transform ◽

Transform Domain ◽

Fusion Method ◽

Nonsubsampled Contourlet Transform ◽

Pulse Coupled Neural Network ◽

Focus Image ◽

Image Fusion Method

Download Full-text

Image Fusion Method Based on Multi-scale Directional Fast Guided Filter and Convolutional Sparse Representation

Recent Advances in Electrical & Electronic Engineering (Formerly Recent Patents on Electrical & Electronic Engineering) ◽

10.2174/2352096511666180607101743 ◽

2019 ◽

Vol 12 (3) ◽

pp. 263-269

Author(s):

Liu Xian-Hong ◽

Chen Zhi-Bin

Keyword(s):

Image Fusion ◽

Sparse Representation ◽

Filter Bank ◽

Guided Filter ◽

Fusion Method ◽

Edge Preserving ◽

Directional Information ◽

Multi Scale ◽

Directional Filter ◽

Image Fusion Method

Background: A multi-scale multidirectional image fusion method is proposed, which introduces the Nonsubsampled Directional Filter Bank (NSDFB) into the multi-scale edge-preserving decomposition based on the fast guided filter. Methods: The proposed method has the advantages of preserving edges and extracting directional information simultaneously. In order to get better-fused sub-bands coefficients, a Convolutional Sparse Representation (CSR) based approximation sub-bands fusion rule is introduced and a Pulse Coupled Neural Network (PCNN) based detail sub-bands fusion strategy with New Sum of Modified Laplacian (NSML) to be the external input is also presented simultaneously. Results: Experimental results have demonstrated the superiority of the proposed method over conventional methods in terms of visual effects and objective evaluations. Conclusion: In this paper, combining fast guided filter and nonsubsampled directional filter bank, a multi-scale directional edge-preserving filter image fusion method is proposed. The proposed method has the features of edge-preserving and extracting directional information.

Download Full-text