The electrophotographic (EP) process widely used in modern laser printer is susceptible to various print defects. Among them, fading is one of the most important ones since it is easily detected by the user on his or her prints; and it occurs frequently i.e. whenever one or more cartridges is low on toner. In contrast, the identification of other types of defects usually requires a specially designed page to be printed and examined by a professionally trained expert. Since fading usually happens when one or more of the printer cartridges – cyan, magenta, yellow or black – is depleted, it is of interest to know which cartridge is depleted given a faded print, so that the user can replace that cartridge. To achieve this goal, we will propose an algorithm that detects color fading from users' prints. The print is scanned; and the resulting digital image is compared to the digital master that was sent to the marking engine to generate the print. This work is a continuation of our previous work, which proposed a text-fading detection algorithm. Here, we extend our work to predict fading in non-text regions as well.