International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 03 | Mar 2024
p-ISSN: 2395-0072
www.irjet.net
MAMDANI AND SUGENO FUZZY INFERENCE SYSTEM BASED MULTIMODAL MEDICAL IMAGE FUSION K.SRIHARI RAO1, CH.RAMBABU2,P.SRI LAKSHMI PRASANNA3, T.SIVA4, P.SHOYAB KHAN5, Sk.Nagulu5, UG Students 1 Associate professor, Department of Electronics and Communication Engineering, Nri institute of technology,
Perechala, Andhra Pradesh,India
2 Associate professor, Department of Electronics and Communication Engineering, Nri institute of technology,
Perechala, Andhra Pradesh,India Department of Electronics and Communication Engineering, Nri institute of technology, Perechala, Andhra Pradesh,India ---------------------------------------------------------------------***--------------------------------------------------------------------3 UG Students,
Abstract - Multimodal medical image fusion is a process of
extracting information from different medical images to obtain a single image called fused image. Fused image analysis is extensively used by clinical professionals for quick diagnosis and treatment of critical diseases. This project is developed using mamdani and sugeno fuzzy inference system for integrated multimodal medical images. Mamdani and sugeno based fusion helps in preservation as well as effective transfer of detailed information present in input images into a fused image. Image Fusion not only provides better information but also minimize the storage cost by minimizing the memory requirement for storage of multiple images. The proposed work is effective and generates better fused images compared to existing techniques such as discrete wavelet transform (DWT) and non-sub sampled counterlet transform (NSCT). The fused image is also compared with quality metrics such as Entropy (E), Mutual Information (MI) and Edge based quality metric (QAB/F). The superiority of the proposed method is presented and is justified using subjective and objective analysis. Key Words: Fuzzy logic, Medical images, Segmentation , Matlab, Image processing.
1.INTRODUCTION The technique of combining all the relevant information from several photos into one or fewer, generally single, images is known as image fusion. By keeping all the data in a single input image rather than using numerous input photos, image fusion reduces storage costs. The primary goal of Image Fusion is to create images that are more suitable and intelligible for both human and machine perception, in addition to minimizing the quantity of data. Image fusion is employed in numerous medical applications, including radiation therapy, neurology, cardiology, and oncology, because of its better and unique information representation. The primary goal of picture fusion is to remove any artifacts while maintaining all relevant and accurate information from the original photos. Various performance measures, such as entropy, correlation coefficient, peak signal to noise © 2024, IRJET
|
Impact Factor value: 8.226
|
ratio, root mean square error, standard deviation, structure similarity index, high pass correlation, edge detection, average gradient et cetera, are used for objective evaluation, which verifies the image quality. The information content of images can be measured by entropy, the registered and fused images can be compared using the correlation coefficient, the fused image's clarity can be assessed using the average gradient, the cumulative error between the original and fused images can be calculated using the root mean square error, and image error can be quantified using the peak signal to noise ratio. Pixel-level, feature-level, and decision-level image fusion algorithms are among the many that have been developed and published in the literature. Based on the brightness of each pixel, pixel level algorithms directly fuse the raw input images. Using their distinguishing characteristics, such as edges and line segments, featurelevel algorithms combine the input images. To create higherquality images, decision level algorithms directly integrate image descriptions, either as relational graphs or as probabilistic variables. Pixel level image fusion methods are more information-efficient than feature-level and decisionlevel algorithms. These pixel-level techniques are more computationally efficient while also being simple to use. For multimodal Image Fusion, pixel-level methods are therefore recommended.
1.1 Motivation Utilizing Mamdani and Sugeno Fuzzy to merge the medical images 1.2 Objectives To improve diagnosis, combine numerous images into one to get essential information. To increase the precision of the diagnosis. To improve the effectiveness of item identification.
2. Literature Survey The fuzzy inference systems of MAMDANI and SUGENO are two well-liked techniques in fuzzy logic for inference and ISO 9001:2008 Certified Journal
|
Page 997