Development of Image Quality Enhancement System for Healthcare Applications

EasyChair Preprint no. 10419

Abstract

To address the presence of noise in the images, the bilateral filter (BF) as a preprocessing step is used in this project. Subsequently, we employed Convolutional Neural Network (CNN) techniques for segmentation to accurately identify the tumor region and enhance an image. Training, testing, and validation of dataset MRI images were used. Based on the technique the images detect tumors or not. The results of the work were evaluated using various performance metrics such as accuracy and SNR. In the proposed methodology image enhancement using Resnet 50 exhibits the superior performance of SNR of 60db and accuracy of 98% by employing image segmentation for the detection of tumor compared to existing approaches.

Keyphrases: Brain Tumor, CNN (Resnet 50), Computed Tomography, Contrast enhancement, image processing, image segmentation, Retinex

@Booklet{EasyChair:10419,
  author = {B.Surekha Reddy and Akhila Gopakumar and Mohith Reddy and Chandana Rachakonda},
  title = {Development of Image Quality Enhancement System for Healthcare Applications},
  howpublished = {EasyChair Preprint no. 10419},

  year = {EasyChair, 2023}}