Research on the Method of Quality Detection of Duck Egg Based on Computer Vision
In order to achieve the intelligence quality detection of duck eggs, the method of weight and yolk color detection of duck eggs based on computer vision was presented. First, based on the maximum entropy principle, the two-dimensional threshold segmentation and mathematical morphology processing were done. Secondly, the relationship between the duck eggs image areas and the weights was established by using polynomial fitting method, and the experiment showed that the detection errors were within ± 2g and the average error was minus 0.13353g. Finally, by using the method of integration of color components, the yolk color feature matrix was constructed, the color moment parameters were extracted from feature matrix and the BP network classifier was designed for classification detection of yolk color. Simultaneously, the overall recognition rate of classifier arrived at 94.30%, indicating that the method is effective.
Zhenyu Du and Bin Liu
Y. Q. Wang et al., "Research on the Method of Quality Detection of Duck Egg Based on Computer Vision", Applied Mechanics and Materials, Vol. 65, pp. 524-529, 2011