This page was created to study and understand the effects of using convolutions (kernels) on images. In simple terms, convolution is the treatment of a matrix (an image in this case) by another one (a filter or a kernel). Here are few things that you can try out here:
Select any image from your device
The size of your image should be at least 10x10
Use filters with sizes 3x3, 5x5 or 7x7
Try out multiple filters, multiple times on a sequence