Fertility Detection of Middle-stage Hatching Egg in Vaccine Production Using Machine Vision

Machine vision was introduced into the hatching egg fertility detection for the production of vaccines, and a method to detect the middle-stage hatching egg fertility by image processing was interpreted. Use the multi-scale morphological transformation to enhance the image, and then detect the shade quality of the image. Next, segment the image local adaptively in order to extract the major embryo blood-vessels of the embryo image, the threshold was counted by histogram-based Weight Fuzzy C-means clustering algorithm. Count the number of the blood-vessels to detect the fertility. The method is experimented on 150 images under the MATLAB7.0. 99.33% images are detected correctly, there is no unfertile eggs being undetected. The execution costs 0.21 seconds on average. The results show that the method is efficient, not sensitive to noise, and it meets the needs of mass production both in detection accuracy rate and executive speed.