Currently, the quality control process in the biscuit industry is carried out in a very rudimentary way and is prone to failures; since it consists of a visual comparison between a sample of the product, which meets previously defined quality standards, with a sample taken at random from a production lot. On this comparison it depends whether the batch goes to the next stages of the process or is discarded._x000D_ The result of the research was a computational vision system that reduces the loss of products in the quality control process in a controlled environment, fulfilling the desired purposes, as well as the implementation of algorithms for image processing of the vision system computational, thus allowing to obtain results with greater accuracy