In image processing and convolutional neural networks (CNNs), convolutions rely on small grids called filters or kernels. These kernels slide across an input image and compute weighted sums, producing what’s known as a feature map. At each position, the kernel’s values are multiplied with the corresponding pixel values beneath it, and the results are added together to form a single output value. This operation repeats as the kernel moves across the image. By adjusting the kernel size or the step