Method and system for identifying individual identities of sheep flock in sheep house based on YOLOv4

Abstract

The invention relates to a YOLOv4 neural network model-based method and system for identifying the individual identity of a sheep flock in a sheep house, and the method comprises the steps: S1, collecting the face information of a sheep, carrying out the preprocessing of a face image, carrying out the labeling of a sheep face through a labeling box, obtaining a data set, and dividing the data set into a training set and a test set; s2, constructing a sheep face recognition neural network model based on YOLOv4, wherein the sheep face recognition neural network model comprises Input, Backbone, Neck and Head; s3, constructing a loss function, carrying out pre-training by using sheep face data, training the loss function by using the training set, taking model parameters obtained after pre-training as initial parameters of the sheep face recognition neural network model, and training the sheep face recognition neural network model by using the training set; and inputting the test set into the trained YOLOv4 neural network model, and evaluating the performance of the neural network model. The method provided by the invention adopts a non-contact identification method, and is low in cost, high in precision, safe and effective, so that the problems of label falling and easy stress of the sheep are avoided.

Description

technical field [0001] The invention relates to the field of modern intelligent animal husbandry, in particular to a method and system for individual identification of sheep in a sheep barn based on a YOLOv4 neural network model. Background technique [0002] In recent years, domestic and foreign animal husbandry is developing from a traditional model to an intelligent, precise and large-scale development. As an important livestock, sheep are an important part of today's animal husbandry. The identification of individual sheep has received extensive attention as an important aspect of large-scale and precise sheep farming. At present, the main sheep identification method in the sheep farming industry is the contact identification method, that is, the method of manual labeling or ear tagging. The most common sheep identification method used in large-scale large-scale farms is the radio frequency-based ear tag method. With the increase of breeding scale, the phenomenon of e...

AI Technical Summary

Problems solved by technology

Cows have a small living density, a small range of activities, and less mutual occlusion, while the breeding density in sheep houses is high, and the amount of activity is large, and mutual occlusion is serious; Liu Bing et al. published a patent "A Yolov4-Based Chicken Farm Raising Chicken Recognition Algorithm" proposed a detection method targeting farmed chickens in farms, but did not identify each individual chicken

Images