Multi-modal image fusion technique for enhancing image quality with multi-scale decomposition algorithm

ABSTRACT

Multi-modality image fusion is an effective technique for fusing complementary information from multi-modality images into an integrated image. The additional information obtained from fused images not only enhances visibility to the human eye, but also complements the limitations of each image. The existing image fusion techniques lacks in preserving the quality of image. Therefore to preserve the structure information and perform the detailed information of source images, a novel image fusion scheme based on Multi-Scale Singular Value Decomposition (MSVD) is proposed. Initially, source images are decomposed into detailed and smooth layer by means of MSVD. Then low- and high-frequency components are extracted with the help of proposed filtering technique and subtraction rule. For efficiently selecting appropriate components, Optimisation algorithm is performed by maximising the mutual information of images. Finally, based on additive fusion rule, the fused images are reconstructed through inverse of MSVD approach. The efficiency of the proposed fusion rule is determined by the comparative analysis with existing image fusion method in terms of Mutual Information (MI), Standard Deviation, Entropy, Correlation etc. The experimental outcome shows the superiority of proposed method and it outperforms state-of-the-art methods by providing better quality image with good clarity. Furthermore, the fused image illustrates better visual effect and edge consistency.

KEYWORDS:

• Fusion
• multi-scale singular value decomposition
• wiener filter
• entropy
• additive rule
• standard deviation
• fusion symmetry
• correlation
• mutual information

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

T.O. Sunitha

T.O. Sunitha completed her UG and PG in the domain of  Computer Applications at Manonmaniam Sundaranar University. She is pursing as a Research Scholar in Department of Computer Applications at Manonmaniam Sundaranar University. Her area of interests is image processing, algorithm techniques, signal processing and communication.

R. Rajalakshmi

Dr.R. Rajalakshmi completed her UG and PG in the domain of  Computer Science at  Noorul Islam College of arts and sciences alnd also completed her PhD in the same domain. Currently she is working as a Assistant  professor and Head of the Department of PG Department  of computer science at Noorul Islam College of arts and sciences. her area of interests is image processing, communication systems.