Design of an artificial vision algorithm to detect cracks in chicken eggs in quality control for poultry companies

Abstract

An algorithm was developed to detect cracks in the external part of the egg applying artificial vision and image processing techniques. The algorithm centers its operation on a Lenovo laptop with an Intel i5 processor and Windows 10 64-bit, which, connected to a cell phone camera with the CAMP STUDIO application, acquires images of the front face of the egg applying the basic ovoscopy technique. The acquired images are processed inside the Lenovo laptop. Statistical and computer vision methods are used to detect abnormalities present on the outside of the egg such as cracks in the shell, excrement, and blood. From the tests carried out with the algorithm, a system efficiency of 96.52% was obtained as a result. It is concluded that the implementation of the algorithm improves the quality and good condition of the egg at the same time allows a greater efficiency of the process, carrying out daily production control in a stored database.