White feather breeding hen group weight estimation system and method based on RGB images

Abstract

The invention discloses a white-feather breeding hen group weight estimation system based on RGB images. The white-feather breeding hen group weight estimation system is characterized by comprising asliding rail type movable device carrying a camera, a wireless communication module, a processing server comprising a white-feather breeding hen individual weight estimation model and a client used for displaying an estimation result. The camera is arranged at the top of the chicken house to shoot a high-angle shot video, and transmits video data to the processing server through the wireless communication module, the processing server calculates the number of breeding hens and estimates the weight of the breeding hens, and an estimation result is displayed by the client. According to the invention, on the basis of image data, non-contact acquisition of phenotypic characteristics of the breeding hens is achieved. The non-contact data acquisition can reduce the interaction between people andchicken flocks, avoid the stress of breeding hens, and improve the welfare of poultry. The network camera is adopted, cost is low, installation is convenient, and application and popularization of the farm are facilitated.

Description

technical field [0001] The invention relates to image processing technology, deep learning technology and wireless network communication technology, which integrates data collection, intermediate transmission, modular processing and graphical interface display, specifically a system for estimating the body weight of white feather breeder chickens based on RGB images and method. Background technique [0002] my country is a big consumer of chicken and eggs, and under the influence of African swine fever, etc., the demand for high-quality chicken and eggs is increasing day by day, and more and more attention has been paid to breeder chickens, the source of high-quality chicken and eggs. Animal weight is one of the main growth indicators concerned by livestock and poultry farming. Due to individual differences, individual animal mass cannot directly reflect its growth status, but the trend of body weight over time reflects the health of animals. The monitoring of the breeder's...

AI Technical Summary

Problems solved by technology

[0005] 1. The depth image has certain restrictions on the shooting angle, which may cause errors;

[0006] 2. For white-feathered breeders, the RGB image features are also distinct. In comparison, the depth image acquisition module is expensive, high in cost, and poor in popularity

Images