Abstract:
In several industries quality control techniques are carried out in order to reduce the amount of products with defects. This increases efficiency and productivity, thus we obtain top quality products and improve business. A biscuit factory has staff to manually separate the cookies that do not meet the standard, they often make mistakes, and therefore consumers find cookies that are different from the regular shape.
Technological and industrial automation have become indispensable tools to prevent and reduce production failures, simplify processes, increase productivity and competitiveness. This has contributed to the development and implementation of automated controls.
In this project we designed an algorithm to detect faults in the production line of a biscuit factory, identifying which products differ from the standard and calculating statistics about these failures. In order to test our algorithm, we used rectangular and triangular cookies.
The algorithm has been developed in MatLab; implementing several Digital Image Processing algorithms, MatLab performed pattern recognition and morphological analysis of parameters such as the cookies’ size and geometric indicators. Throughout a graphical interface we can depict the final results either by cookie or as
