Eggshell Deformation Detection Applying Computer Vision

This work describes a method for detecting deformations on curved surfaces, focusing especially on the study of deformations in poultry eggs to identify shell damages caused by thin-shelled eggs, or in coltsfoot. The proposed system is based on the detection of defects by sweeping a pattern of laser structured light and imaging, highlighting the changes in geometry as a result of the laser transitions deformation generated by scanning the surface of egg; then the images are analyzed to obtain equidistant points along the curve and evaluated by generating a cubic spline interpolation that allows the extraction of descriptive metric characteristics to observe the disparity between curves showing deformation by performing a graph interposition, the metric information acquired is used to classify the defective samples developing an artificial neural network, achieving a 97.5% efficiency during evaluation.