About this document
Convolution and Edge Detection Techniques by subbu2051 is a document available to read on EtoBox.
1) Gaussian filtering smooths an image by applying a Gaussian kernel that gives less weight to pixels further from the center. This reduces noise and detail. 2) Convolving an image with derivative filters like the Laplacian of Gaussian (LoG) detects edges by finding zero crossings in the second derivative. The LoG approximates the ideal edge detection operator. 3) Blurring an image with a Gaussian filter removes high frequency details like edges before subtracting the blurred version from the original, enh
- Author
- subbu2051
- Language
- EN