Study on Automatic Shape Identification of Hatching Eggs Based on an Improved GA Neural Network

Shape inspection of hatching eggs is an important and hard work in farms. Manual inspection lacks of objectivity and is time-consuming. An automatic shape identifying method was proposed based on machine vision, moment technique and improved GA-NN algorithm. Egg form index and radius differences are extracted as eggs shape feature parameters. An improved immune GA algorithm is put forward to optimize topology structure of LMBP-NN. After identified egg shape index, radius differences are used as inputs of LMBP-NN and its outputs are used to determine the hatching eggs shape normal or not. Its classification accuracy is reached to 97.1% for longer eggs, 95.59% for shorter eggs, 94.87% for abnormal eggs and 95.75% for normal eggs respectively. It is significant for shape identification of hatching eggs automatically, which could improve detection accuracy and efficiency. The neural network system for shape identification of hatching eggs has a high accuracy and generalization ability, and the algorithm is feasible and robust.