Every vision model, from a simple filter to a large neural network, starts with the same raw material: an image. Before training anything, it helps to know what an image actually is to a computer, how a camera captures it, and how to load and change it in code.
A digital image is a grid of pixels, each holding numbers that describe brightness or color. Those numbers are stored in channels: one for a grayscale image, three for a color image. A camera sensor records them through a pattern of color filters, and color spaces such as RGB, HSV, and grayscale are different ways of describing the same picture.
In practice you work with images as arrays. NumPy holds the pixel data and OpenCV reads, resizes, converts, and saves images. Watch the lessons in order, then continue on to Classical Computer Vision.
An introduction to how images are represented digitally, starting from pixels and the values stored in them.
How a camera sensor captures an image and records color using a Bayer filter pattern.
Color versus grayscale images, the RGB, HSV, and HSL color spaces, and how to convert between them.
A hands-on OpenCV course in Python: reading images, resizing, color spaces and channels, and the other basics you will use throughout this course.