Unlock instant, AI-driven research and patent intelligence for your innovation.

Image-based cattle uniqueness recognition method, device and management system

A recognition method and image technology, applied in biometric recognition, neural learning methods, character and pattern recognition, etc., can solve the problems of low effectiveness of cattle and low detection accuracy of cattle face recognition, and achieve accurate recognition

Pending Publication Date: 2022-02-18
四川楠水农牧科技有限公司 +1
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The technical problem solved by the present invention is that the prior art has low detection accuracy for cow face recognition and the effectiveness of current cattle supervision is low. The purpose of the present invention is to provide an image-based unique cattle recognition method, device and management system, the present invention extracts the characteristics of the nose pattern of the cow for identification and detection, so as to achieve accurate identification of the individual cow, puts the identity information of the cow after identification into the cow body recognition system, and can grasp the actual situation of the cow in real time

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Image-based cattle uniqueness recognition method, device and management system
  • Image-based cattle uniqueness recognition method, device and management system
  • Image-based cattle uniqueness recognition method, device and management system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0045] Such as figure 1 As shown, a kind of image-based cattle uniqueness identification method that this embodiment one provides, the method comprises:

[0046]In step S1, the bullnose pattern image is preprocessed, and the processed image is sent to the convolutional neural network to extract the feature quantity of the bullnose pattern, and the registration of the bullnose pattern is carried out;

[0047] In step S2, collect the bullnose pattern that needs to be identified, preprocess the collected bullnose pattern image, send the preprocessed image to the deep learning model for feature extraction, and obtain the feature vector;

[0048] In step S3, the feature vector is matched with the registered feature quantity, and the cosine similarity is calculated;

[0049] In step S4, a threshold is set, and the cosine similarity is compared with the threshold to obtain a cow recognition result.

[0050] Specifically, in step S1, such as figure 2 As shown, the specific steps o...

Embodiment 2

[0064] The second embodiment provides a cow identification device on the basis of the first embodiment, the device includes:

[0065] The first acquisition module preprocesses the bullnose pattern image, sends the processed image to the convolutional neural network to extract the feature quantity of the bullnose pattern, and registers the bullnose pattern;

[0066] The second acquisition module collects the bullnose pattern that needs to be identified, preprocesses the collected bullnose pattern image, sends the preprocessed image to the deep learning model for feature extraction, and obtains the feature vector;

[0067] A calculation module, configured to match the feature vector with the registered feature quantity, and calculate the cosine similarity;

[0068] The identification module is used to set a threshold, compare the cosine similarity with the threshold, and obtain the identification result of the cow.

[0069] Preferably, the first acquisition module and the secon...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an image-based cattle uniqueness recognition method and device and a management system, and relates to the field of cattle nose print recognition. The method comprises the steps of pre-processing a cattle nose print image, sending the processed image into a convolutional neural network to extract the characteristic quantity of the cattle nose prints, and registering the cattle nose prints; acquiring the cattle nose prints needing to be recognized, pre-processing the acquired cattle nose print image, and sending the pre-processed image into a deep learning model for feature extraction to obtain a feature vector; matching the feature vector with the feature quantity, and calculating the cosine similarity of the feature vector and the feature quantity; and setting a threshold value, and comparing the cosine similarity with the threshold value to obtain a cattle recognition result. According to the method, the extracted features are matched with the cattle nose prints registered in a database to finally obtain the matched cattle nose prints, so that the cattle individuals are accurately recognized.

Description

technical field [0001] The invention relates to the field of cattle identification and detection, more specifically, it relates to an image-based unique identification method, device and management system for cattle. Background technique [0002] At present, the digitization and intelligence of my country's agriculture and animal husbandry have just started, especially small and medium-sized pastures, cooperatives and individual farmers lack effective means. Traditional supervision manually records and calculates the number of cattle, whether they are stolen, daily entry and exit, and body condition, which may easily lead to data errors and affect the effectiveness of supervision, making it difficult to grasp the situation of cattle in a timely manner. Therefore, it is urgent to solve the pain points related to agriculture and animal husbandry through intelligent means. Taking animal husbandry supervision as the starting point, making innovations and breakthroughs, overcomi...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06V40/10G06V10/40G06V10/75G06V10/74G06V10/82G06K9/62G06N3/04G06N3/08G06F16/51G06F16/532G06F16/583G06Q50/02
CPCG06N3/08G06F16/51G06F16/532G06F16/583G06Q50/02G06N3/047G06N3/045G06F18/22
Inventor 谈耀易军王巍甘佳贺芳方东辉石溢邓小东阿果约达李榜东张俤张鹏
Owner 四川楠水农牧科技有限公司