Egg Grading Algorithm Based On Color Clustering

Egg grading manual Produce Tremendous amount of work, the test results do not have the objectivity and accuracy, and can not meet the needs of modern industrial production. Global Egg automatic grading technology based on computer vision technology and based on acoustics and optical technology. Acoustic and optical technology used in the egg grading equipment costs high and damage in the cause eggshell. Based on computer vision technology is more suitable for egg processing industry in China to promote.The topic based on computer vision technology for egg grading, using color clustering algorithm to analyze the image, measured according to Haugh clustering index and egg shape index. Color clustering algorithm is based on clustering algorithm and the specific color space model for the cluster analysis. Commonly used clustering algorithm is divided into color threshold method and regional approach. Threshold method because of its stable performance and simple becomes a basic technology of color clustering.In the traditional clustering algorithm, K-means algorithm and fuzzy C means clustering algorithm are better. Evidential C means algorithm based on the fuzzy C means algorithm Improves credible division. The introduction of modern heuristic algorithms traditional clustering algorithm can make up for the shortcomings of traditional clustering and deficiencies, genetic algorithms, neural networks and ant colony algorithm combined with the traditional clustering algorithm, the clustering effect is remarkable.This paper design egg grading algorithm based on genetic algorithm and design evidential C means algorithm, divided into clustering algorithm, freshness grading of eggs algorithm and quality grading of eggs algorithm. The clustering algorithm is the focus contents of this paper.Egg clustering algorithm use HIS color model, binary coding based on cluster center based, initial population by H component charts, design adaptive operator for genetic manipulation, Evidential C means algorithm optimization each generation of populations, ultimately the best cluster center matrix optimum solution, and then decode the output clustering.According to egg clustering algorithm, analysis Haugh index and egg shape index, achieve freshness grading of eggs and quality grading of eggs. The experiment proved that the use of histogram for initial population clustering can improve the effectiveness. ECM optimization can improve the accuracy of clustering, genetic operators designed to improve the convergence performance. Haugh index and egg shape index determination algorithm for high accuracy. The Experiment Also proved that the algorithm is a practical guide to production practice.