Selection of image fusion quality measures: objective, subjective, and metric assessment

Dixon, Timothy D.; Canga, Eduardo Fernandez; Nikolov, SG; Trościanko, Tom; Noyes, Jan; Canagarajah, C N; Bull, David

doi:10.1364/josaa.24.00b125

Cited by 17 publications

(13 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on their solutions, our proposed solution improves in the representation of detailed information, and furtherly enhances the performance of the fused image. The objective evaluation of the fused image includes the AG (Average Gradient) [75], Edge-Intensity (EI) [76], Edge Retention Q AB/F [77], Mutual Information (MI) [78], and Visual Information Fidelity (VIF) [79]. In this paper, the size of the learned dictionary is set to 128, and the size of the image patch is set to 8 × 8.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

An Integrated Dictionary-Learning Entropy-Based Medical Image Fusion Framework

Wang

Zhang

et al. 2017

Future Internet

View full text Add to dashboard Cite

Abstract:Image fusion is widely used in different areas and can integrate complementary and relevant information of source images captured by multiple sensors into a unitary synthetic image. Medical image fusion, as an important image fusion application, can extract the details of multiple images from different imaging modalities and combine them into an image that contains complete and non-redundant information for increasing the accuracy of medical diagnosis and assessment. The quality of the fused image directly affects medical diagnosis and assessment. However, existing solutions have some drawbacks in contrast, sharpness, brightness, blur and details. This paper proposes an integrated dictionary-learning and entropy-based medical image-fusion framework that consists of three steps. First, the input image information is decomposed into low-frequency and high-frequency components by using a Gaussian filter. Second, low-frequency components are fused by weighted average algorithm and high-frequency components are fused by the dictionary-learning based algorithm. In the dictionary-learning process of high-frequency components, an entropy-based algorithm is used for informative blocks selection. Third, the fused low-frequency and high-frequency components are combined to obtain the final fusion results. The results and analyses of comparative experiments demonstrate that the proposed medical image fusion framework has better performance than existing solutions.

show abstract

Section: Methodsmentioning

confidence: 99%

“…• EI measures the strength of image local changes [76]. A higher EI value implies the fused image contains more intensive edges.…”

Section: Image Quality Comparisonmentioning

confidence: 99%

An Integrated Dictionary-Learning Entropy-Based Medical Image Fusion Framework

Wang

Zhang

et al. 2017

Future Internet

View full text Add to dashboard Cite

show abstract

“…The DT-CWT has been adapted by Nikolov and colleagues [41] to enable it to be used for image fusion. When used in human assessment, the DT-CWT has been shown to perform better than the DWT in subjective comparison [41], as well as outperforming other fusion methods in a range of objective and subjective assessment tasks [42][43][44].…”

Section: Dual-tree Complex Wavelet Transformmentioning

confidence: 99%

“…In the current context, the use of subjective ratings tasks is not viable, as they are particularly time-consuming and troublesome with video assessment processes. Moreover, it is worth noting that subjective ratings have been shown to lead to differing patterns of results from objective human tasks [42][43][44], making their value in real terms questionable.…”

Section: Assessment Of Fused Imagesmentioning

confidence: 99%

See 1 more Smart Citation

Task-based scanpath assessment of multi-sensor video fusion in complex scenarios

et al. 2010

Self Cite

View full text Add to dashboard Cite

Fusion of noisy images based on joint distribution model in dual‐tree complex wavelet domain

Sun

Zhu

Luo

et al. 2018

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Source images are frequently corrupted by noise before fusion, which will lead to the quality decline of fused image and the inconvenience for subsequent observation. However, at present, most of the traditional medical image fusion scheme cannot be implemented in noisy environment. Besides, the existing fusion methods scarcely make full use of the dependencies between source images. In this research, a novel fusion algorithm based on the statistical properties of wavelet coefficients is proposed, which incorporates fusion and denoising simultaneously. In the proposed algorithm, the new saliency and matching measures are defined by two distributions: the marginal statistical distribution of single wavelet coefficient fit by the generalized Gaussian Distribution and joint distribution of dual source wavelet coefficients modeled by the anisotropic bivariate Laplacian model. Additionally, the bivariate shrinkage is introduced to develop a noise robust fusion method, and a moment-based parameter estimation applied in the fusion scheme is also exploited in denoising method, which allows to achieve the consistency of fusion and denoising. The experiments demonstrate that the proposed algorithm performs very well on both noisy and noise-free images from multimodal medical datasets (computerized tomography, magnetic resonance imaging, magnetic resonance angiography, etc.), outperforming the conventional methods in terms of both fusion quality and noise reduction. K E Y W O R D Sbi-shrinkage, image fusion, KL-divergence, mutual information, statistical distribution 1 | INTRODUCTION Imagology plays an important role in medical science, owing to the application for the visualization of anatomy structure, function, and metabolism information. 1 Actually, singular modality medical images exhibit limitations due to the imaging mechanism. Computerized tomography (CT) images can provide bone tissue information with high resolutions, but they cannot show the soft tissue. Although magnetic resonance imaging (MRI) images are good at showing the anatomical information of soft tissue, they cannot provide metabolism information, while T2 weighted MRI (MRI-T2) and magnetic resonance angiography (MRA) images could achieve good vision of pathological tissue. 2 In practical clinical applications, the multi-modal medical image fusion technology, which combines the complementary information from multi-modal images, is helpful to facilitate a better observation of anatomical and metabolism information. 3 This technology provides an easy but very efficient access for diagnosis and assessment of disease. 4 Image fusion based on multiscale transform has become a more common approach for pixel-level image fusion and abundant research results have been achieved. For waveletbased fusion methods, wavelet transforms are performed first on each source image to obtain different sub-bands which contain decomposition information of varying

show abstract

Selection of image fusion quality measures: objective, subjective, and metric assessment

Cited by 17 publications

References 19 publications

An Integrated Dictionary-Learning Entropy-Based Medical Image Fusion Framework

An Integrated Dictionary-Learning Entropy-Based Medical Image Fusion Framework

Task-based scanpath assessment of multi-sensor video fusion in complex scenarios

Fusion of noisy images based on joint distribution model in dual‐tree complex wavelet domain

Contact Info

Product

Resources

About