Maximizing Insights from MRI Brain Images Segmentation through HSV Histogram and Gabor Wavelet Transform, and Machine Learning-Assisted Image Retrieval

Medical imaging has better managed disease and reduced patient suffering by detecting brain tumours. Medical professionals employ magnetic resonance imaging (MRI). Due to its non-invasiveness and high resolution and contrast, MRI can see tissues. MRI may take several images of the same tissue region with improved contrast. Many photographs provide additional anatomical information about the same tissue. Data from many contrast mechanisms can help researchers understand brain illness. Because tumour tissue varies in location, size, shape, and texture, feature extraction is difficult. MR brain imaging components including white matter, grey matter, and cerebrospinal fluid have complicated structures that make feature extraction difficult. The literature on feature extraction for tumour segmentation does not compare which method is best. Here, we implement two commonly used libraries of established and successful texture feature extraction algorithms. By collecting frequency, localization, and orientation, Gabor wavelet feature extraction delivers multi-resolution surface information in both spatial and frequency domains. These feature extraction algorithms reflect the intensity relationship between image pixels or clusters of pixels. They may also predict image quality using first- and second-order statistics. Our research examines the efficacy and complexity of these two feature extraction algorithms for tumour diagnosis and segmentation. The researchers employed a locust-based genetic algorithm to analyse MR images perimage, reducing the likelihood that the results were just machine-Iearning- based.