Egg counting and positioning system in stacked cage culture mode

Abstract

The invention discloses an egg counting and positioning system in a stacked cage culture mode, and the system comprises an egg target tracking and counting module which achieves the counting of eggs based on an egg multi-category target detection model and a multi-target tracking DeepSORT algorithm, and distinguishes different categories of eggs to achieve the respective counting; and the speed display screen positioning and reading recognition module is used for designing and training a speed display screen semantic segmentation model based on a UNet algorithm and a speed display screen target classification model based on a CNN algorithm by utilizing a convolutional neural network, and the speed display screen semantic segmentation model and the speed display screen target classification model are matched to jointly realize recognition of reading of the speed display screen so as to obtain the operation speed of the egg belt. According to the system, the target tracking algorithm is applied to egg counting in the stacked cage culture mode for the first time, the situation of repeated counting is avoided, and the counting accuracy is improved.

Description

technical field [0001] The technical field of the present invention is the field of livestock and poultry precision breeding, and the technologies used mainly include: deep learning technology, image processing technology and network communication technology, the research object is eggs in a stacked cage mode, and particularly relates to a stacked cage Egg counting and positioning system in rearing mode. Background technique [0002] Eggs are a common food in daily life. They have the advantages of rich nutrition, reasonable price and wide circulation. The market demand for them is large. The quantity ranks first in the world. In order to meet the needs of the consumer market, laying hen farming is developing in the direction of large-scale, modern, intensive, and full-scale mechanization. Among them, the application of stacked cage equipment is becoming more and more extensive. The automatic egg collection system equipped with such equipment includes: Vertical egg collect...

AI Technical Summary

Problems solved by technology

[0003] In the current production practice of laying hens, machine vision technology is mainly used to detect the apparent characteristics of eggs for egg identification; the counting of eggs is mainly solved by installing counting components at the connection between the chain conveyor belt and the longitudinal elevator to solve the problem of multiple egg adhesion , or through image processing to binarize the collected images to achieve counting, but the above methods are not conducive to the positioning of eggs

No reports on egg counting and positioning in stacked cages

Images