چكيده به لاتين
Analysis of the glomerulus’ geometry is important in histopathological evaluation of renal microscopic images. Due to shape and size disparity of even the glomeruli of the same person, automatic detection and analysis of these renal objects is not an easy task. Although manual measurements are time consuming and are not accurate, but it is commonly being used in medical centers. In this research, a new method based on Fourier transform is proposed to detect these objects and their geometrical parameters. So, five descriptors based on Hough transform, circle fitting, and ellipse fitting are suggested to distinguish the regions of interest. Images of 6 Wistar rats consist our dataset, and their glomerulus’ and Bowman’s space are calculated. The results are verified by a renal pathologist, and maximum error of 6.26% is resulted in calculation of glomeruli’s areas. Having tested different glomeruli with various shapes, the experimental consequences showed robustness and reliability of our method. Therefore, it could be used to show renal diseases and glomerular disorders by measuring morphological changes both accurately and quickly.