Check patentability & draft patents in minutes with Patsnap Eureka AI!

Cattle body size algorithm based on deep learning and feature part detection

A technology of deep learning and position, applied in the field of cattle body measurement algorithm, can solve the problem of inability to measure the body size of cattle safely and accurately, and achieve the effect of reducing measurement cost, reducing labor intensity and improving efficiency

Pending Publication Date: 2019-08-16
INNER MONGOLIA UNIV OF SCI & TECH +1
View PDF11 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the problem that the body size of cattle cannot be safely and accurately measured in the prior art, the present invention provides a cattle body size algorithm based on deep learning and characteristic part detection

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
  • Cattle body size algorithm based on deep learning and feature part detection
  • Cattle body size algorithm based on deep learning and feature part detection
  • Cattle body size algorithm based on deep learning and feature part detection

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0040] A cow body measurement algorithm based on deep learning and characteristic part detection, comprising the following steps:

[0041] Step 1, set up a measurement system, please refer to figure 2, which is a schematic diagram of the overall structure of a preferred embodiment of a cow body measurement algorithm based on deep learning and characteristic part detection in the present invention, including a ground label surface 1, a sideways camera 2, a tail camera 3 and an image server, please refer to figure 1 , is a schematic diagram of the circuit principle of a preferred embodiment of the cow body measurement algorithm based on deep learning and characteristic part detection in the present invention, the sideways camera 2 collects the cow's sideways image and transmits it to the image server, and the tail camera 3 collects the cow's body image The tail image is transmitted to the image server, and the image server includes a feature part detection model and a body meas...

Embodiment 2

[0076] Taking the cow's head on the left side of the cow's tail as an example, the category of the cow's hooves can be judged by the following steps:

[0077] (1) Because the bull's head is on the left side of the cow's tail, according to the pixel coordinate position, draw a dotted line from the center of the bull's head to the center of the cow's tail, and from the bull's head to the tail are the front hoof (①②) and the rear hoof (③④);

[0078] (2) According to the first step, the front hoof and the rear hoof have been judged, and the head of the middle bull is on the left side of the tail, and the hoof of the cow whose landing point is close to the front reference line according to the ground marking surface is the left front hoof of the cow (that is, ① is the left front hoof of the cow) , and the other is the left hind hoof of the cow (that is, ③ is the left hind hoof of the cow); on the contrary, the reference line near the rear side is the right front hoof (that is, ② is ...

Embodiment 3

[0080] The present embodiment has enumerated following five kinds of calculation methods of body height of cattle:

[0081] method one:

[0082] (1) Using the bovine torso feature box

[0083] (2) if figure 1 As shown in , obtain the feature frame height H1 of the cow torso, which is the pixel height of the cow.

[0084] (3) Select the calibration parameters of the plane area where the cattle are located, and use the actual body height equal to the pixel body height multiplied by the calibration parameters to obtain the actual body height S.

[0085] (4) The calculation formula is as follows:

[0086] S=H1*K n (K n is the calibration parameter of the plane where the cow is located)

[0087] Method Two:

[0088] (1) The bovine torso feature box and the hoof feature box are used.

[0089] (2) if figure 2 As shown, the coordinates (X1, Y1) of the corner point of the feature frame of the torso of the cow and the coordinates (X2, Y2) of the midpoint of the two front hoofs...

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 provides a cattle body size algorithm based on deep learning and feature part detection, and the algorithm comprises the steps: 1, setting a measurement system which comprises a ground marking surface, a side camera, a tail camera, and an image server, wherein the image server comprises a characteristic part detection model and a body size measurement algorithm; 2, using the side camera and the tail camera to collect a side image and a tail image of the cow respectively, transmitting the images to the image server, using a characteristic part detection model to process an input image and outputting characteristic part coordinate information; 3, calculating a ground calibration parameter An and a calibration parameter Kn of a calibration surface vertical to the ground; and 4,obtaining body size data of a cow according to the coordinate information and the proportion. The side camera and the tail camera are adopted to collect the images of the cattle, the image server processes the image data to calculate the body size data of the cattle, the labor intensity and difficulty of measurement are reduced, and potential safety hazards are avoided.

Description

technical field [0001] The invention relates to the technical field of cattle body measurement algorithms, in particular to a cattle body measurement algorithm based on deep learning and characteristic part detection. Background technique [0002] With the development of science and technology and the advent of the era of big data, the informatization and intelligence of animal husbandry is the top priority to promote the rapid and healthy development of animal husbandry in my country. With the help of intelligent and automated data collection and computer-aided analysis, the production scale of animal husbandry can be improved, labor costs can be reduced, and production efficiency can be enhanced. [0003] The body size indicators of cattle in the animal husbandry industry mainly include parameters such as height, body length, body oblique length, chest width, hindquarter width, and chest girth length. With the development of biological research and the continuous accumula...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/62G06T7/73G06K9/00G06K9/62
CPCG06T7/62G06T7/73G06T2207/20081G06T2207/20084G06V40/103G06F18/24
Inventor 李琦赵建敏白卓玉张万锴杜永兴尚绛岚
Owner INNER MONGOLIA UNIV OF SCI & TECH
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com