한우의 3차원 영상에서 결정된 몸통 체적을 이용한 체중 추정 = Estimation of Body Weight Using Body Volume Determined from Three-Dimensional Images for Korean Cattle

국문 초록 (Abstract)

가축의 체중은 사료 요구량과 영양 상태를 평가하는 데 필요한 주요 지표에 해당한다. 본 연구는 한우의 3-D 영상으로부터 몸통 체적을 산출한 후 체중을 추정하고자 시도되었다. 한우의 3-D 영상 획득에 640×480 픽셀의 해상도, 44fps의 프레임속도 및 47°(H)×37°(V)의 화각을 갖는 TOF 카메라가 사용되었다. 획득된 3-D 영상에서 배경과 몸통 분리, 이상치 제거 등의 전처리 과정을 거쳐서 몸통에 대한 격자 영상을 얻었다. 또한 각각의 격자에 깊이 정보를 적용한 수치적분으로 몸통 체적을 결정하였다. Calibration dataset에서 체중과 몸통체적의 선형회귀에 대한 결정계수는 0.8725로 나타났다. 한편 몸통 체적에 월령을 설명 변수로 추가한 체중 추정의 중회귀 모형에서 결정계수는 0.9083으로 나타났다. Validation dataset에서 중회귀 모형을 이용한 체중 추정의 MAPE와 RMSE는 각각 8.2%, 24.5kg으로 나타났다. 결과적으로 체중 추정을 위한 회귀 모형의 성능이 개선되고, 체중 추정에 소요되는 노력이 절감됨을 고려한다면 3-D 영상에서 결정된 몸통 체적이 한우의 체중 추정에 유효한 변수로 사용될 것이다.

다국어 초록 (Multilingual Abstract)

Body weight of livestock is a crucial indicator for assessing feed requirements and nutritional status. This study was performed to estimate the body weight of Korean cattle (Hanwoo) using body volume determined from three-dimensional (3-D) image. A TOF camera with a resolution of 640×480 pixels, a frame rate of 44 fps and a field of view of 47°(H)×37°(V) was used to capture the 3-D images for Hanwoo. A grid image of the body was obtained through preprocessing such as separating the body from background and removing outliers from the obtained 3-D image. The body volume was determined by numerical integration using depth information to individual grid. The coefficient of determination for a linear regression model of body weight and body volume for calibration dataset was 0.8725. On the other hand, the coefficient of determination was 0.9083 in a multiple regression model for estimating body weight, in which the age of Hanwoo was added to the body volume as an explanatory variable. Mean absolute percentage error and root mean square error in the multiple regression model to estimate the body weight for validation dataset were 8.2% and 24.5kg, respectively. The performance of the regression model for weight estimation was improved and the effort required for estimating body weight could be reduced as the body volume of Hanwoo was used. From these results obtained, it was concluded that the body volume determined from 3-D of Hanwoo could be used as an effective variable for estimating body weight.

참고문헌 (Reference)

1 서광욱, "젖소 체중추정을 위한 영상처리 알고리즘" 한국농업기계학회 36 (36): 48-57, 2011

2 Radu Bogdan Rusu, "Towards 3D Point cloud based object maps for household environments" Elsevier BV 56 (56): 927-941, 2008

3 Kawasue K., "Three-dimensional shape measurement system for black cattle using KINECT sensor" 7 : 222-230, 2013

4 S. Ozkaya, "The relationship of parameters of body measures and body weight by using digital image analysis in pre-slaughter cattle" Copernicus GmbH 51 (51): 120-128, 2008

5 Lee M. Y., "Relation and estimation heritabilities for body weight and body measurements of Korean cattle (Hanwoo)" 27 : 691-695, 1985

6 M. Haile-Mariam, "Prediction of liveweight of cows from type traits and its relationship with production and fitness traits" American Dairy Science Association 97 (97): 3173-3189, 2014

7 A.J. Heinrichs, "Predicting Body Weight and Wither Height in Holstein Heifers Using Body Measurements" American Dairy Science Association 75 (75): 3576-3581, 1992

8 Lvwen Huang, "Non-Contact Body Measurement for Qinchuan Cattle with LiDAR Sensor" MDPI AG 18 (18): 3014-, 2018

9 Jiahua Wu, "Extracting the three-dimensional shape of live pigs using stereo photogrammetry" Elsevier BV 44 (44): 203-222, 2004

10 장동화, "Estimation of Body Weight for Korean Cattle Using Three-Dimensional Image" 한국농업기계학회 45 (45): 325-332, 2020

11 Kramer O, "Dimensionality reduction with unsupervised nearest neighbors" Springer 13-22, 2013

12 Roii Spoliansky, "Development of automatic body condition scoring using a low-cost 3-dimensional Kinect camera" American Dairy Science Association 99 (99): 7714-7725, 2016

13 Nabil Brandl, "Determination of live weight of pigs from dimensions measured using image analysis" Elsevier BV 15 (15): 57-72, 1996

14 Sakir Tasdemir, "Determination of body measurements on the Holstein cows using digital image analysis and estimation of live weight with regression analysis" Elsevier BV 76 (76): 189-197, 2011

15 S. Viazzi, "Comparison of a three-dimensional and two-dimensional camera system for automated measurement of back posture in dairy cows" Elsevier BV 100 : 139-147, 2014

16 Kashiha M., "Automatic weight estimation of individual pigs using image analysis" 107 : 38-44, 2014

17 M.F. Hansen, "Automated monitoring of dairy cow body condition, mobility and weight using a single 3D video capture device" Elsevier BV 98 : 14-22, 2018

18 Song X., "Automated body weight prediction of dairy cows using 3-dimensional vision" 101 : 4448-4459, 2018

19 Mateus P. Gionbelli, "Achieving Body Weight Adjustments for Feeding Status and Pregnant or Non-Pregnant Condition in Beef Cows" Public Library of Science (PLoS) 10 (10): e0112111-, 2015

20 Andrea Pezzuolo, "A Feasibility Study on the Use of a Structured Light Depth-Camera for Three-Dimensional Body Measurements of Dairy Cows in Free-Stall Barns" MDPI AG 18 (18): 673-, 2018

21 장동화, "3차원 영상을 이용한 원예산물의 크기와 플러그묘의 평균초장 추정" (사) 한국생물환경조절학회 28 (28): 358-365, 2019

22 Miguel Garrido, "3D Maize Plant Reconstruction Based on Georeferenced Overlapping LiDAR Point Clouds" MDPI AG 7 (7): 17077-17096, 2015

23 Manuel Vázquez-Arellano, "3-D Imaging Systems for Agricultural Applications—A Review" MDPI AG 16 (16): 618-, 2016