An Efficient Biomedical Color Image Retrieval System Based on Continuous Orthogonal Legendre Fourier Quaternion

Human volunteers must be used in studies for biomedical research in order to improve knowledge of health-related issues, particularly when it comes to medical color image diagnosis. An extraction system's effectiveness is determined by how well it can recover features using a feature descriptor. Among two kinds of medical images, the quaternion continuous orthogonal Legendre Fourier moments (QCOLFMs) outperform quaternion orthogonal Fourier Mellin moments (QOFMMs) and quaternion radial associated Laguerre moments (QRALMs) in terms of feature extraction. Our work entails evaluating the normalized image reconstruction error (NIRE) values of color medical images in relation to the average retrieval precision (AP), the average retrieval rate (AR), and another average measure ARP. This has important ramifications for magnetic resonance imaging (MRIs) and other medical scanners. The medical scanners and MRIs databases offer a brief summary of the used image retrieval systems, and the proposed moments (QCOLFMs) strategy yields better performance and results than other strategies like moments QOFMMs and moments QRALMs.