Detection of Corn Leaf Grey Spots Using Coefficient Bounds Obtained for a Subclass of Analytic Functions

An essential area of research in computer vision is edge detection. Edge detection is a technique that aims to describe the intensity variations in an image in terms of the underlying physical processes. The recognition and characterisation of noticeable intensity variations is a critical objective of edge detection. The Corn Leaf dataset, comprising of 574 images, is used to test the suggested method, which is based on the convolution of coefficient bounds with the image pixels determined for a subclass \(\mathfrak {R}(t,\delta )\) that is subordinate to the petal-shaped domain. The PSNR, SSIM, MSE, RMSE, and DSSIM metrics are examined to check the quality of the enhanced images. The proposed edge detector is estimated for eight directions in the \(3\times 3\) neighbourhood. Calculations have been made for all eight convolution masks and also the average of the angles for the edge detection operators.