基于深度图像的多姿态肉牛体尺自动测量方法

doi:10.12133/j.smartag.SA202210001

Smart Agriculture ›› 2022, Vol. 4 ›› Issue (4): 144-155.doi: 10.12133/j.smartag.SA202210001

基于深度图像的多姿态肉牛体尺自动测量方法

叶文帅^1,²(), 康熙^2,³, 贺志将^1,², 李孟飞^1,², 刘刚^1,²()

^1.中国农业大学智慧农业系统集成研究教育部重点实验室，北京 100083
^2.中国农业大学农业农村部农业信息获取技术重点实验室，北京 100083
^3.浙大宁波理工学院计算机与数据工程学院，浙江宁波 315000

收稿日期:2022-10-06 出版日期:2022-12-30
基金项目:
国家重点研发计划项目(2021YFD1300502)
作者简介:叶文帅（1997－），男，硕士研究生，研究方向为农业关键信息技术。E-mail：yewenshuai@cau.edu.cn
通信作者: 刘刚（1966－），男，博士，教授，研究方向为电子信息技术农业应用。E-mail：pac@cau.edu.cn

Automatic Measurement of Multi-Posture Beef Cattle Body Size Based on Depth Image

YE Wenshuai^1,²(), KANG Xi^2,³, HE Zhijiang^1,², LI Mengfei^1,², LIU Gang^1,²()

^1.Key Lab of Smart Agriculture Systems, Ministry of Education, China Agricultural University, Beijing 100083, China
^2.Key Laboratory of Agricultural Information Acquisition Technology, Ministry of Agriculture and Rural Affairs, China Agricultural University, Beijing 100083, China
^3.School of Computing and Data Engineering, NingboTech University, Ningbo 315200, China

Received:2022-10-06 Online:2022-12-30
Foundation items:National Key Research and Development Program of China (2021YFD1300502)
About author:YE Wenshuai, E-mail：yewenshuai@cau.edu.cn
Corresponding author:LIU Gang， E-mail：pac@cau.edu.cn

摘要/Abstract

摘要：

养殖场中肉牛较为活跃，采集得到的图像数据中肉牛姿态多变，肉牛姿态端正帧较少，导致自动测量肉牛体尺困难。针对以上问题，本研究通过分析肉牛骨架特征和肉牛图像边缘轮廓特征，提出一种多姿态肉牛体尺自动测量方法。首先，利用深度相机Azure Kinect DK从正上方采集肉牛俯视深度视频数据，对视频数据进行分帧处理；其次，对原始深度图像进行预处理，将肉牛从复杂的背景中提取出来；再次，利用Zhang-Suen算法提取目标图像肉牛骨架，检测骨架交点和端点，分析肉牛头部特征，并确定头部去除点，去除图像中肉牛头部信息；最后，利用改进的U弦长曲率算法提取肉牛轮廓曲率曲线，根据曲率值确定体尺测点，将体尺测点转换到三维空间中，计算体尺参数。本研究通过分析大量深度图像数据，将图像中肉牛姿态分为左歪、右歪、姿态端正、低头和抬头五类。试验结果表明，本研究提出的基于骨架的多姿态肉牛头部去除方法在5种姿态下的头部去除成功率均高于92%；在23头肉牛不同姿态共46帧深度图像中，利用基于改进U弦长曲率的体尺测点提取方法，测得体直长测量的平均绝对误差为2.73 cm，体高测量的平均绝对误差为2.07 cm，腹宽测量的平均绝对误差为1.47 cm。研究结果可为精确测量多姿态下肉牛体尺提供支撑。

关键词: 肉牛体尺测量, 深度图像, 多姿态, Zhang-Suen算法, 改进的U弦长曲率算法

Abstract:

Beef cattle in the farm are active, which leads the collection of posture of the beef cattle changeable, so it is difficult to automatically measure the body size of the beef cattle. Aiming at the above problems, an automatic measurement method for beef cattle's body size under multi-pose was proposed by analyzing the skeleton features of beef cattle head and the edge contour features of beef cattle images. Firstly, the consumer-grade depth camera Azure Kinect DK was used to collect the top-view depth video data directly above the beef cattle and the video data were divided into frames to obtain the original depth image. Secondly, the original depth image was processed by shadow interpolation, normalization, image segmentation and connected domain to remove the complex background and obtain the target image containing only beef cattle. Thirdly, the Zhang-Suen algorithm was used to extract the beef cattle skeleton of the target image, and calculated the intersection points and endpoints of the skeleton, so as to analyze the characteristics of the beef cattle head to determine the head removal point, and to remove the beef cattle head information from the image. Finally, the curvature curve of the beef cattle profile was obtained by the improved U-chord curvature method. The body measurement points were determined according to the curvature value and converted into three-dimensional spaces to calculate the body size parameters. In this paper, the postures of beef cattle, which were analyzed by a large amount of depth image data, were divided into left crooked, right crooked, correct posture, head down and head up, respectively. The test results showed that the head removal method proposed based on the skeleton in multiple postures hads head removel success rate higher than 92% in the five postures. Using the body measurement point extraction method based on the improved U-chord curvature proposed, the average absolute error of body length measurement was 2.73 cm, the average absolute error of body height measurement was 2.07 cm, and the average absolute error of belly width measurement was 1.47 cm. The method provides a better way to achieve the automatic measurement of beef cattle body size in multiple poses.

Key words: beef body size measurement, depth image, multi-gesture, body size measurement, Zhang-Suen algorithm, improved U-chord curvature body

中图分类号:

TP391

叶文帅, 康熙, 贺志将, 李孟飞, 刘刚. 基于深度图像的多姿态肉牛体尺自动测量方法[J]. 智慧农业(中英文), 2022, 4(4): 144-155.

YE Wenshuai, KANG Xi, HE Zhijiang, LI Mengfei, LIU Gang. Automatic Measurement of Multi-Posture Beef Cattle Body Size Based on Depth Image[J]. Smart Agriculture, 2022, 4(4): 144-155.

参考文献

1	Food and Agriculture Organization of the United Nations, FAOSTAT[EB/OL].[2022-02-17]. .
2	李孟飞, 康熙, 王彦超, 等. 基于阈值分析的曲线拟合和边缘检测生猪侧视点云补全[J]. 农业机械学报, 2021, 52(S1): 81-88.
2	LI M, KANG X, WANG Y, et al. Side-view point cloud completion method combining cubic b-spline curve fitting and edge delection based on threshold analysis[J]. Transactions of the CSAM, 2021, 52(S1): 81-88.
3	FREITAG G P, LIMA L G F, JACOMINI J A, et al. An accurate image analysis method for estimating body measurements in horses[J]. Journal of Equine Veterinary Science, 2021, 101: ID 103418.
4	SPOLIANSKY R, EDAN Y, PARMET Y, et al. Development of automatic body condition scoring using a low-cost 3-dimensional Kinect camera[J]. Journal of Dairy Science, 2016, 99(9): 7714-7725.
5	NIR O, PARMET Y, WERNER D, et al. 3D Computer-vision system for automatically estimating heifer height and body mass[J]. Biosystems Engineering, 2018, 173: 4-10.
6	SALAU J, HAAS J H, JUNGE W, et al. A multi-Kinect cow scanning system: Calculating linear traits from manually marked recordings of Holstein-Friesian dairy cows[J]. Biosystems Engineering, 2017, 157: 92-98.
7	赵建敏, 赵成, 夏海光. 基于Kinect v4的牛体尺测量方法[J]. 计算机应用, 2022, 42(5): 1598-1606.
7	ZHAO J, ZHAO C, XIA H. Cattle body size measurement method based on Kinect v4[J]. Journal of Computer Application, 2022, 42(5): 1598-1606.
8	李嘉位, 马为红, 李奇峰, 等. 复杂环境下肉牛三维点云重建与目标提取方法[J]. 智慧农业(中英文), 2022, 4(2): 64-76.
8	LI J, MA W, LI Q, et al. Automatic Acquisition and target extraction of beef cattle 3D point cloud from complex environment[J]. Smart Agriculture, 2022, 4(2): 64-76.
9	赵新强. 基于Kinect的奶牛体尺检测与试验研究[D]. 泰安: 山东农业大学, 2018.
9	ZHAO X. Measurement and test of dairy cattle body size based on Kinect[D]. Taian: Shandong Agricultural University, 2018.
10	常海天. 种牛体尺非接触测量系统研究[D]. 长春:长春工业大学, 2018.
10	CHANG H. The study on the non-cantact measurement system of cattle body[D]. Changchun: Changchun University of Technology, 2018.
11	司永胜, 安露露, 刘刚, 等. 基于Kinect相机的猪体理想姿态检测与体尺测量[J]. 农业机械学报, 2019, 50(1): 58-65.
11	SI Y, AN L, LIU G, et al.Ideal posture detection and body size measurement of pig based on Kinect[J]. Transactions of the CSAM, 2019, 50(1): 58-65.
12	ZHANG L, ZHOU W, LI J, et al. Histogram of oriented gradients feature extraction without normalization[C]// 2020 IEEE Asia Pacific Conference on Circuits and Systems (APCCAS). Piscataway, New York, USA: IEEE, 2020: 252-255.
13	张文飞, 韩建海, 郭冰菁, 等. 改进的分水岭算法在粘连图像分割中的应用[J]. 计算机应用与软件, 2021, 38(6): 243-248.
13	ZHANG W, HAN J, GUO B, et al. Application of improved watershed algorithm in adhesion image segmentation[J]. Computer Applications and Software, 2021, 38(6): 243-248.
14	ZHANG T, SUEN C. A fast parallel algorithm for thinning digital patterns[J]. Communications of the ACM, 1984, 27(3): 236-239.
15	郭娟娟. 基于U弦长曲率的碎片拼接算法[D]. 苏州: 苏州大学, 2014.
15	GUO J. An algorithm for fragments re-assembly based on U-chord curvature[D]. Suzhou: Soochow University, 2014.
16	牛金玉. 基于三维点云的奶牛体尺测量与体重预测方法研究[D]. 杨凌 :西北农林科技大学, 2018.
16	NIU J. Body size measurement and weight prediction for dairy cows based on 3D point cloud[D]. Yangling: Northwest A&F University, 2018.
17	张馨月. 基于深度相机的生猪体尺自动测量技术研究[D]. 北京: 中国农业大学, 2021
17	ZHANG X. Research on automatic measurement technology of pig body size based on depth camera[D]. Beijing: China Agricultural University, 2021.
18	RUCHAY A, KOBER V, DOROFEEV K, et al. Accurate body measurement of live cattle using three depth cameras and non-rigid 3-D shape recovery[J]. Computers and Electronics in Agriculture, 2020, 179: ID 105821.
19	李琦, 刘伟, 赵建敏. 基于双目视觉及Mask RCNN的牛体尺无接触测量[J]. 黑龙江畜牧兽医, 2020(6): 46-50, 159-160.
19	LI Q, LIU W, ZHAO J. Non-contact measurement of bovine body size based on bincular vision and Mask RCNN[J]. Heilongjiang Animal Science and Veterinary Medicine, 2020(6): 46-50, 159-160.

[1]	代昕, 王军号, 张翼, 王鑫杰, 李晏兴, 戴百生, 沈维政. 基于时空流特征融合的俯视视角下奶牛跛行自动检测方法[J]. 智慧农业(中英文), 2024, 6(4): 18-28.
[2]	翁智, 范琦, 郑志强. 基于多模态图像信息及改进实例分割网络的肉牛体尺自动测量方法[J]. 智慧农业(中英文), 2024, 6(4): 64-75.

基于深度图像的多姿态肉牛体尺自动测量方法

Automatic Measurement of Multi-Posture Beef Cattle Body Size Based on Depth Image

在线阅读

知网下载

本地下载

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

编辑推荐

Metrics

本文评价