Abstract This work presents the development of an application designed in Python using the Kivy and KivyMD libraries (for cross-platform) to identify the result of two six-sided dice in an environment with uncontrolled lighting to amplify the visual adequacy of people with low or no visual capacity; as an environment, we propose the use of a template similar to the casino game known as Craps. We structure the developed software in four stages. The first stage consists of linking the APP with a digital camera either connected to a local network through an IP address or wired via USB; the second stage consists of determining the right time to process the image of interest through the acquisition and comparison of features such as color and position of the dice; the third stage seeks to perform the segmentation, and non-binary classification of the value reported by each die through intelligent computer vision algorithms such as color space conversion, Euclidean distance, and CNN; and the last stage presents the results obtained after the classification through a loudspeaker to inform people with no visual ability and through enlarged digital text for people with low visual capability. The possible expected result is a system that integrates the four stages mentioned above through an intuitive and accessible low-cost APP designed in Python, aimed primarily at people with low or no visual capability. Due to the characteristics of board games that make use of 6-sided dice, people with low or no visual capacity face a problem of quality or interpretation of results after a pair of dice have been thrown, which generates a situation of indirect exclusion by inhibiting this part of the population from accessing this type of recreation or social interaction regularly.