International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 185 - Number 24 |
Year of Publication: 2023 |
Authors: Deepak B. Patil, T.S. Vishwanath |
10.5120/ijca2023923000 |
Deepak B. Patil, T.S. Vishwanath . Specular Reflection Analysis and Image Denoising in Cervix Images. International Journal of Computer Applications. 185, 24 ( Jul 2023), 33-37. DOI=10.5120/ijca2023923000
Colposcopy is a crucial imaging technique for finding cervical abnormalities. Colposcopic image analysis, particularly the accurate segmentation of the cervical region, has significant clinical significance. It is suggested to divide and extract the cervical region in a medical and anatomical sense using a cervical segmentation approach based on the HSV colour mode. First, the histogram (Y) of the colposcopic image is examined using the histogram threshold approach. Pretreatment is required in the colposcopy image in order to remove the mirror reflection. The Preprocessed RGB pictures are used second. Segmentation of the cervical region can be obtained by applying the area filter to smooth the edge. 100 common colposcopy photos that had been expertly tagged were evaluated and verified for our study. Accuracy, sensitivity, specificity positive predictive value, and negative predictive value were used to analyse and compare the findings. Our experimental findings demonstrate that the method's accuracy, specificity, and sensitivity are, respectively, 88.25%, 80.99%, and 95%.