The primary purpose of convolutions is to reduce the dimensions of an image (width, height, and number of channels). The larger the image, the more processing time is required.
The strides parameter causes a kernel to skip over pixels in an image and not include them in the output. The strides parameter determines how a convolution operation works with a kernel when a larger image and more complex kernel are used. As a convolution is sliding the kernel over the input, it is using the strides parameter to determine how it walks over the input, instead of going over every element of an input.
Let's take a look at the following example, where we are moving a 3 x 3 x 1 kernel over a 6 x 6 x 1 image with a stride of 1, 3, 3, 1:
Step 1 ...
