Estimation of Body Weight for Korean Cattle Using Three-Dimensional Image
- PDF / 2,075,220 Bytes
- 8 Pages / 595.276 x 790.866 pts Page_size
- 15 Downloads / 183 Views
Online ISSN 2234-1862 Print ISSN 1738-1266
ORIGINAL ARTICLE
Estimation of Body Weight for Korean Cattle Using Three-Dimensional Image Dong Hwa Jang 1,2 & Chulsoo Kim 3,4
&
Yeoung-Gyu Ko 5 & Yong Hyeon Kim 3,4
Received: 21 August 2020 / Revised: 5 October 2020 / Accepted: 9 November 2020 # The Korean Society for Agricultural Machinery 2020
Abstract Purpose This study was conducted to estimate the body weight of Korean cattle (Hanwoo) using three-dimensional image. Methods A time-of-flight camera and a stereo vision camera were used to acquire top-view images for Hanwoo. Multiple linear regression model incorporating body dimensions and age was developed to estimate the body weight. Results The prediction models with or without age variable were determined. Incorporating age variable that correlates with body weight improved the coefficient of determination for both TOF images and stereo vision images. The weight estimation error was greatly reduced under the condition that calves less than 6 months old were excluded. Conclusions It was concluded that 3-D image is a promising technique for estimating the body weight of Hanwoo cattle. Keywords Body weight . Korean cattle . Point cloud data . 3-D camera . 3-D image
Introduction Body weight of cattle is a key factor in assessing the growth performance (Fox et al. 1999), productivity (Gionbelli et al. 2015), selection, and breeding (Pilarczyk and Wojcik 2008). The periodical measurement of body weight is used to evaluate the growth response to nutrient supply and health anomalies (Kawasue et al. 2013). Some farmers use a measuring tape or a weighing scale with steel frame to measure the body
* Yong Hyeon Kim [email protected] 1
Department of Agricultural Machinery Engineering, School of Graduate, Jeonbuk National University, Jeonju 54896, Republic of Korea
2
Present address: Animal Environment Division, National Institute of Animal Science, Wanju-gun 55365, Republic of Korea
3
Department of Bioindustrial Machinery Engineering, College of Agriculture & Life Sciences, Jeonbuk National University, Jeonju 54896, Republic of Korea
4
Institute for Agricultural Machinery & ICT Convergence, Jeonbuk National University, Jeonju 54896, Republic of Korea
5
Animal Genetics Resources Research Center, National Institute of Animal Science, Hamyang-gun 50000, Republic of Korea
weight of Korean cattle (Hanwoo). However, this work can be labor intensive for the farmer as well as stressful for the cattle. Thus, farmers often rely on their visual judgment in estimating the body weight, which is a subjective method whose accuracy depends on their experience. Computer vision system has been widely used in engineering. A number of applications have also been reported in animal science (Kashiha et al. 2014; Ozkaya and Bozkurt 2008; Seo et al. 2011; Tasdemir et al. 2011). The advantage of these methods is non-contact and non-destructive measurement. However, most of these applications are limited to twodimensional (2-D) analyses and thus the accuracy of the results is not so high
Data Loading...
