An edge filter will produce an image where all the edges of the input image are highlighted. This is done by looking for pixels that have high contrasts with their surrounding neighbours. You may have guessed that we can do this using convolutions. In fact, for each pixel, we will perform two convolutions, one to look for contrast in the horizontal direction and the other for the vertical direction. This convolution does not need to be divided by the sum of kernels because we are not averaging. The result is a vector in the X and Y direction and so we want to calculate the magnitude of that vector for our final result. There are a few different kernels we can use to do this but one of the most popular is the Sobel Kernel. Below is an example of how the calculation is done. Sometimes the magnitude will be larger than $255$ and so in that case we have to cap it at $255$ and if it is less than $0,$ set it to $0.$