Design And Performance Of An Automatic Egg Sorting System Based On Computer Vision

An automatic system for eggs sorting by indirect weight and shape assessment using computer vision is presented in the paper. Some basic geometric parameters of the eggs, namely: minor and major axis, area and perimeter, shape factor and shape index, are obtained using image processing algorithms. The weight is assessed from the geometric parameters using a regression model. The sorting accuracy of the proposed system is evaluated for two transport conveyors speeds, corresponding to 2 and 3 eggs per second. It is found that the overall sorting accuracy in each case is 94.6% and 90.3% respectively.