A pixel (picture element) is the smallest addressable element in a raster image or the smallest unit of a digital image that can be displayed and represented on a digital display device. Each pixel has a value (or values for multiple channels) that typically represent color or intensity. For example, in a grayscale image, a pixel’s value might be 0-255 representing darkness to brightness. In a color image (e.g., RGB), each pixel actually consists of three components (red, green, blue values). The resolution of an image is given in pixels (width x height). Pixels are the basic input for many computer vision algorithms; a 256x256 image is essentially an array of 65,536 pixels, each with some intensity values. When zooming into an image on a computer, you eventually see the individual blocky pixels. Processing images often entails manipulating pixel values (filtering, etc.). In deep learning for vision, pixel values are often normalized (e.g., scaled between 0 and 1 or mean-subtracted) and then fed into conv nets.