The process of convolution emphasizes a pattern of interest in a particular image by processing it with another smaller image called a filter (also called a kernel). For example, if we want to find the edges of objects in an image, we can convolve the image with a particular filter to get them. Edge detection can help us in object detection, object classification, and other applications. So, the process of convolution is about finding characteristics and features in an image. 

The approach to finding patterns is based on finding patterns that can be reused on different data. The reusable patterns are called filters or kernels.