Vision-based chick male and female intelligent identification method and system

Abstract

The invention provides a vision-based chick male and female intelligent identification method and system, and the method comprises the steps: obtaining and collecting a plurality of images at the anus of a chick through an optical imaging module, and forming a chick male and female image data set; enabling the chick male and female identification module to identify the network model according to the chick male and female image data set and then perform multi-task model learning according to the basic identification network model to obtain N chick male and female identification models for N chicken breeds; enabling the optical imaging module to collect an image of the anus of a to-be-detected chick and inputting the image into the chick male and female identification module, and enabling the chick male and female identification module to judge and output a male and female result according to the chick male and female identification model of the corresponding chicken breed. The method and system have the advantages that the male and female chicks are distinguished through the identification model, the human eye distinguishing function is replaced, and the human eyes are liberated; the accuracy is improved, the labor cost is reduced, and the working efficiency is greatly improved.

Description

technical field [0001] The invention relates to a vision-based intelligent identification method and system for male and female chicks. Background technique [0002] Male and female feeding are carried out separately in the farms. The main reason is that the farms divide laying hens and broiler chickens. Of course, only hens can be raised in laying hen farms; although commercial broilers have both male and female chickens, if they are mixed, due to the development of cocks Fast, and it will grab food, which will affect the growth of hens, so it is best to keep them separately to ensure economic benefits. Therefore, early gender identification of chicks can eliminate roosters or separate male and female chicks, which saves various feeding costs such as feed, poultry houses, and labor. [0003] At present, the fast and slow feather identification method and the anal identification method are mainly used to distinguish between male and female chicks. [0004] The fast and slo...

AI Technical Summary

Problems solved by technology

The main disadvantage of this method is that the feathers of newborn chicks are not fully developed, and the accuracy of the fast and slow feather method is low; the fast and slow feather method is only useful for specific types of chickens, and the method is not universal

[0008] Because the difference between male and female features is very small, it is difficult to train the discriminator. It usually takes three months to learn the basics, and the identification accuracy can reach more than 95% after one year; the shape of the male and female feature points is very small, and it needs to be under strong light. It can only be seen, and long-term observation is harmful to the human eye; the accuracy of manual classification is not high, generally around 95%; the speed of manual classification is not fast, generally 1000-1200 per hour, which is difficult to meet the production needs of enterprises; artificial fatigue is easy , the accuracy and speed will drop significantly after work

Images