Edge Detection in Digital Images Through Fast Gradient Filters using Fuzzy Inference System
Rakesh Ranjan, Vinay Avasthi · 2022 IEEE World Conference on Applied Intelligence and Computing (AIC) · 2022
Digital image processing is required for the analysis of an image. To perform the analysis different types of feature extraction approaches are introduced. The approaches work pretty good and perform better analysis in some areas. But many times, these extractions of features approaches fail to perform well when there is an injection of noise into the images. The work under this article resolves this issue by dealing with the noisy image. Our work mainly focuses on the edge detection of digital images using fuzzy logic using Fast Guided Filter. The contrast of the image is used to compute the threshold of the image which is accountable in the case of the noisy image. Edge maps from the threshold images are produced by the fast-guided filters and the fuzzy inference system. The Root Mean Square Error (RMSE) and Peak Signal to Noise Ratio (PSNR) parameters are calculated for measuring the edge accuracy at different noise levels. The proposed model has achieved an RMSE of 0.12 and PSNR of 57.31.