Detection and recognition method for sheep movement behavior in grazing in grassland, and device thereof

A recognition method and behavior technology, applied in the field of detection, can solve the problems of low detection and recognition accuracy and high labor intensity, and achieve the effect of improving the accuracy and enhancing the denoising ability.

Inactive Publication Date: 2019-01-11
INNER MONGOLIA UNIVERSITY
View PDF1 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Aiming at the deficiencies of the prior art, the present invention provides a detection and recognition method and device for the motion behavior of grazing sheep in grasslands, which has the advantages of convenience and accuracy, and solves the problem of low accuracy in the detection and recognition of motion behaviors of grazing sheep in grasslands in the prior art. labor-intensive problem

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
  • Detection and recognition method for sheep movement behavior in grazing in grassland, and device thereof
  • Detection and recognition method for sheep movement behavior in grazing in grassland, and device thereof
  • Detection and recognition method for sheep movement behavior in grazing in grassland, and device thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0073] During the detection process of the movement behavior data of grazing sheep collected by the three-axis acceleration sensor, the signal disturbance (noise) will be generated due to the fixed position of the device installation and the diversity of sheep activity behavior. In summary, there are the following types of noise: (1) Detect white noise that is disturbed in the hardware circuit; (2) Noise caused by body shaking caused by emotional fluctuations caused by grazing sheep due to fright. The existence of noise directly affects the quality of the original signal, resulting in the inaccuracy of motion behavior detection, so it is imperative to denoise the motion behavior data of grazing sheep.

[0074] In the field of denoising research, most researchers use Fourier transform to denoise data. The Fourier transform converts the time-domain signal into a frequency-domain signal, so that the low-frequency signal or high-frequency signal passes through the algorithm, so as...

Embodiment 2

[0158] In Embodiment 1, due to the discontinuity of the hard threshold function, there will be a large oscillation at a certain point when denoising the acceleration data signal, resulting in a denoising effect that cannot meet the requirements; while the soft threshold function is between λ and - λ is continuous and has a good denoising effect. However, due to the processing of subtracting the threshold value during the operation, some signals are distorted, and the reconstructed acceleration data signal will have a certain attenuation.

[0159] The denoising effects of different threshold functions are quite different. There will always be some oscillations in the denoising process of the hard threshold function, while the denoising method of the soft threshold function has continuous characteristics and shows better smoothness. There will be a phenomenon of data distortion after filtering. By calculating the signal-to-noise ratio of each wavelet denoising, the following co...

Embodiment 3

[0166] This embodiment discloses a detection and recognition device for the motion behavior of grazing sheep in grassland, which adopts the detection and recognition method for motion behavior of grazing sheep in grassland in Embodiment 1, and the detection and recognition device includes a three-axis acceleration sensor, a wavelet denoising module, a preprocessing module, cluster processing module.

[0167] The three-axis acceleration sensor is installed on the grazing sheep, and obtains the movement behavior signal of the grazing sheep;

[0168] The wavelet denoising module performs wavelet decomposition on the acceleration data of the noise-containing grazing sheep movement behavior, selects a wavelet threshold, and inversely transforms and reconstructs the effective acceleration data signal;

[0169]The preprocessing module performs position calibration and windowing processing on the motion behavior signal; the position calibration formula is:

[0170] A x =a x sinθsin...

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 a detection and recognition method for sheep movement behavior in grazing in grassland, and a device thereof. The detection and recognition method comprises the following steps: firstly, carrying out wavelet decomposition on the acceleration data of the grazing sheep motion behavior with noise, then a wavelet threshold is selected, and finally the effective acceleration data signal is reconstructed by inverse transform; 2, carrying out position calibration and windowing processing on the motion behavior signal; Step 3, selecting K-Means clustering algorithm, the collected three-axis acceleration data samples are assigned to K clusters, and recalculating the new clustering positions trained by each iteration, so that the sum of distances from each data to the new clustering center point is minimized, and the distance from each sample point to the clustering center satisfies the error sum of squares minimum termination condition, and the loop is jumped out by repeating the steps. The invention reduces the influence of noise on the motion behavior signal of grazing sheep and improves the accuracy of recognition data through denoising, preprocessing and clustering processing of the motion behavior signal.

Description

technical field [0001] The invention relates to a detection and recognition method and device for motion behavior in the technical field of detection, in particular to a detection and recognition method and device for the motion behavior of grazing sheep on grasslands. Background technique [0002] my country is particularly rich in grassland resources. The total area of ​​grassland in the country is nearly 6 billion mu, accounting for about 41.7% of the total area of ​​my country. In order to maintain regional and ecological balance, grassland resources play an extremely important role. Grassland resources are the basis of grazing. Grazing activities directly affect grassland ecosystems. Grazing livestock not only eat grassland plants, but also trample plants by walking, which will seriously affect grassland vegetation and grassland resources. [0003] With the development of animal welfare breeding and intelligent animal husbandry, the detection and identification of anima...

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): A61B5/11A61B5/00G06K9/00G06K9/62
CPCA61B5/1118A61B5/1121A61B5/7203A61B5/7235G06F2218/04G06F18/23
Inventor 韩丁刘艳秋王鑫张旭温军苏赫武佩杜德鹏马子寅
Owner INNER MONGOLIA UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap