Method and system for facial identification of animals under small sample conditions

Abstract

The invention relates to a method and system for facial identification of animals under small sample conditions. The method comprises steps that features of pixels corresponding to each preset coordinate in a to-be-identified animal facial image are extracted, and feature vectors of the to-be-identified animal facial image are obtained according to the features of the pixels corresponding to eachpreset coordinate; the feature vectors of the to-be-identified animal facial image are inputted to a preset classifier, according to the output result of the preset classifier, the identification result of an animal face in the to-be-identified animal facial image is obtained. The method is advantaged in that the original feature vectors are expanded, the classifier is trained through obtaining alarger number of sample feature vectors, the training effect of the preset classifier obtained through the training is better, and accuracy of identifying individual animals in the to-be-identified animal facial image is improved according to the preset classifier in the case of the small sample size.

Description

technical field [0001] The embodiments of the present invention relate to the technical field of computer vision, and in particular to an animal face recognition method and system under the condition of small samples. Background technique [0002] Computer vision and intelligent monitoring have gradually penetrated into various fields of the breeding industry. For example, in the refined breeding of dairy cows, the establishment of individual files, information collection and traceability of dairy products all require fast and accurate individual identification of dairy cows. The traditional cow identification method writes or prints the cow number on the ear tag, and then installs the ear tag on the cow's ear by puncturing. Traditional ear tags need to be read by human eyes, and their simplicity is limited by light and distance, and they are prone to failure due to damage and stains. With the development of intelligent sensing technology, RFID electronic ear tags are used...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the problem of insufficient accuracy of animal face recognition under small sample conditions in the prior ar

Images