基于多角度Kinect v2的羊只三维模型重构方法研究

doi:10.13733/j.jcam.issn.2095-5553.2024.03.026

摘要/Abstract

摘要： 羊只的体尺参数是衡量其生长发育状况、生产性能和遗传特性的关键指标。重建羊只的三维模型可以为自动化获取多种羊只体尺参数提供数据基础，因此提出一种基于多角度Kinect v2的羊只三维模型重构方法。该方法通过放置在羊只顶部和左右两个侧面的Kinect v2设备，获取羊只的三维点云数据；利用这些数据中的点云之间的相对位置关系，进行点云坐标的转换和初始配准；采用ICP算法进行精确配准建立三维模型。结果表明：当Kinect v2深度相机高度为120 cm、俯视角为30°时，获取的点云质量较高，自动配准的平均误差为0.233 cm，平均耗时为12.89 s。根据模型计算出的羊只体高、体斜长、十字部高和腰脚宽等体尺参数与实际测量平均误差均在5%以内。

关键词: 羊只, Kinect v2, 点云配准, 三维重建, 体尺计算

Abstract: The body size parameters of sheep are the key indicators to assess the growth, performance and genetic characteristics. The reconstruction of the 3D model can provide data basis for automatic acquisition of various body size parameters for sheep. In this paper, a reconstruction method is designed for the 3D model of sheep based on multiangle Kinect v2. By placing Kinect v2 devices on the top and both sides of the sheep, threedimensional point cloud data of the sheep can be captured. Then, by utilizing the relative positional relationships between the point clouds in the data, coordinate transformation and initial alignment of the point clouds are performed. Finally, the Iterative Closest Point (ICP) algorithm is used for precise registration to establish the threedimensional model. The results show that when the height of Kinect v2 depth camera is 120 cm and the overlooking angle is 30°, the quality of the obtained point cloud is higher. The average error of automatic registration is 0.233 cm, and the average time of automatic registration is 12.89 s. The average error of body size parameters such as body height, body oblique length, cross height and waist and foot width calculated by the model and the actual measurement are within 5%.

Key words: sheep, Kinect v2, point cloud registration, threedimensional reconstruction, body size calculation

中图分类号:

范才虎, 程曼, 袁洪波, 王媛, 蔡振江. 基于多角度Kinect v2的羊只三维模型重构方法研究[J]. 中国农机化学报, 2024, 45(3): 189-197.

Fan Caihu, Cheng Man, Yuan Hongbo, Wang Yuan, Cai Zhenjiang. Reconstruction method of the 3D model for sheep based on multiangle Kinect v2[J]. Journal of Chinese Agricultural Mechanization, 2024, 45(3): 189-197.

参考文献

［1］　Matte A, Waquil P D. Changes in markets for lamb in livestock family farming in Brazil ［J］. Small Ruminant Research, 2021, 205: 106535.
［2］　Gurgel A L C, dos Santos Difante G, Neto J V E, et al. Prediction of dry matter intake by meat sheep on tropical pastures ［J］. Tropical Animal Health and Production, 2021, 53: 1-8.
［3］　Mahon N, Clark B, Proctor A, et al. Exploring farmers understanding of and responses to endemic animal health and welfare issues in the UK ［J］. Veterinary Record, 2021, 189(10): 1-8.
［4］　Chikwanha O C, Mupfiga S, Olagbegi B R, et al. Impact of water scarcity on dryland sheep meat production and quality: Key recovery and resilience strategies ［J］. Journal of Arid Environments, 2021, 190: 104511.
［5］　Pallottino F, Steri R, Menesatti P, et al. Comparison between manual and stereovision body traits measurements of Lipizzan horses ［J］. Computers and Electronics in Agriculture, 2015, 118: 408-413.
［6］　Rathbun F M, Pralle R S, Bertics S J, et al. Relationships between body condition score change, prior midlactation phenotypic residual feed intake, and hyperketonemia onset in transition dairy cows ［J］. Journal of Dairy Science, 2017, 100(5): 3685-3696.
［7］　马学磊, 薛河儒, 周艳青, 等. 基于改进区域生长法的羊体点云分割及体尺参数测量［J］. 中国农业大学学报, 2020, 25(3): 99-105.
Ma Xuelei, Xue Heru, Zhou Yanqin, et al. Point cloud segmentation and measurement of the body size parameters of sheep based on the improved region growing method ［J］. Journal of China Agricultural University, 2020, 25(3): 99-105.
［8］　孙芊芊, 李海军, 宣传忠, 等. 基于羊只应激反应的智能饲喂机器人功能与造型研究［J］. 内蒙古农业大学学报（自然科学版）, 2019, 40(5): 60-64.Sun Qianqian, Li Haijun, Xuan Chuanzhong, et al. Research on function and modeling of intelligent feeding robot based on sheep stress response ［J］. Journal of Inner Mongolia Agricultural University (Natural Science Edition), 2019, 40(5): 60-64.
［9］　Buddle E A, Bray H J, Ankeny R A. “Of course we care!”: A qualitative exploration of Australian livestock producers understandings of farm animal welfare issues ［J］. Journal of Rural Studies, 2021, 83: 50-59.
［10］　杨鹏. 羊常见传染病的预防及治疗方法［J］. 畜牧兽医科技信息, 2021(9): 138.
［11］　Marinello F, Pezzuolo A, Cillis D, et al. Application of KinectSensor for threedimensional body measurements of cows ［J］. Proceedings of the 7th European Precision Livestock Farming, ECPLF, 2015.
［12］　Zhou Y, Xue H, Wang C, et al. Reconstruction and body size detection of 3d sheep body model based on point cloud data ［C］. IFIP WG International Conference on Computer and Computing Technologies in Agriculture, 2019: 251-262.
［13］　Wang K, Guo H, Ma Q, et al. A portable and automatic Xtionbased measurement system for pig body size ［J］. Computers and Electronics in Agriculture, 2018, 148: 291-298.
［14］　Condotta I C F S, BrownBrandl T M, SilvaMiranda K O, et al. Evaluation of a depth sensor for mass estimation of growing and finishing pigs ［J］. Biosystems Engineering, 2018, 173: 11-18.
［15］　Alvarez J R, Arroqui M, Mangudo P, et al. Body condition estimation on cows from depth images using Convolutional Neural Networks ［J］. Computers and Electronics in Agriculture, 2018, 155: 12-22.
［16］　Zhang A L, Wu B P, Wuyun C T, et al. Algorithm of sheep body dimension measurement and its applications based on image analysis ［J］. Computers and Electronics in Agriculture, 2018, 153: 33-45.
［17］　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.
Si Yongsheng, An Lulu, Liu Gang, et al. Ideal posture detection and body size measurement of pig based on Kinect ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(1): 58-65.
［18］　郝雪萍. 基于图像处理的杜泊羊体重估算模型研究［D］. 武汉: 武汉理工大学, 2015.
［19］　张丽娜, 杨建宁, 武配, 等. 羊只形态参数无应激测量系统设计与试验［J］. 农业机械学报, 2016, 47(11): 307-315.
Zhang Lina, Yang Jianning, Wu Pei, et al. Design and experiment of nonstress measuring system for sheeps conformation parameters ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2016, 47(11): 307-315.
［20］　Khojastehkey M, Aslaminejad A A, Shariati M M, et al. Body size estimation of new born lambs using image processing and its effect on the genetic gain of a simulated population ［J］. Journal of Applied Animal Research, 2016, 44(1): 326-330.
［21］　Vieira A, Brando S, Monteiro A, et al. Development and validation of a visual body condition scoring system for dairy goats with picturebased training ［J］. Journal of Dairy Science, 2015, 98(9): 6597-6608.
［22］　Shi C, Teng G, Li Z. An approach of pig weight estimation using binocular stereo system based on LabVIEW ［J］. Computers and Electronics in Agriculture, 2016, 129: 37-43.
［23］　Yang L, Wang B, Zhang R, et al. Analysis on location accuracy for the binocular stereo vision system ［J］. IEEE Photonics Journal, 2017, 10(1): 1-16.
［24］　Cui Y, Zhou F, Wang Y, et al. Precise calibration of binocular vision system used for vision measurement ［J］. Optics Express, 2014, 22(8): 9134-9149.
［25］　赵建敏, 赵忠鑫, 李琦. 基于Kinect传感器的羊体体尺测量系统［J］. 传感器与微系统, 2015, 34(9): 100-103.Zhao Jianmin, Zhao Zhongxin, Li Qi. Sheep body size measurement system based on Kinect sensor ［J］. Transducer and Microsystem Technologies, 2015, 34(9): 100-103.
［26］　Pezzuolo A, Gonzàlez L A, Giora D, et al. Body measurements of dairy cows using a structure from motion (SfM) photogrammetry approach ［C］. 8th European Conference on Precision Livestock Farming, ECPLF， 2017: 483-492.
［27］　Pezzuolo A, Milani V, Zhu D H, et al. Onbarn pig weight estimation based on body measurements by structurefrommotion (SfM) ［J］. Sensors, 2018, 18(11): 3603.
［28］　初梦苑, 刘刚, 司永胜, 等. 基于三维重建的奶牛体重预估方法［J］. 农业机械学报, 2020, 51(S1): 378-384.
Chu Mengyuan, Liu Gang, Si Yongsheng, et al. Predicting method of dairy cow weight based on threedimensional reconstructions ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51(S1): 378-384.
［29］　张馨月, 刘刚, 经玲, 等. 基于点云精简的奶牛背部体尺测点自动提取方法［J］. 农业机械学报, 2019, 50(S1): 267-275.
Zhang Xinyue, Liu Gang, Jing Ling, et al. Automatic extraction method of cows back body measuring point based on simplification point cloud ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(S1): 267-275.
［30］　蔡静怡, 严飞, 吴兆祥, 等. 基于Kinect v2的多视角三维重建实现［J］. 传感技术学报, 2020, 33(8): 1149-1154.Cai Jingyi, Yan Fei, Wu Zhaoxiang, et al. Implementation of multiview 3D reconstruction based on Kinect v2 ［J］. Chinese Journal of Sensors and Actuators, 2020, 33(8): 1149-1154.
［31］　Le Cozler Y, Allain C, Caillot A, et al. Highprecision scanning system for complete 3D cow body shape imaging and analysis of morphological traits ［J］. Computers and Electronics in Agriculture, 2019, 157: 447-453.
［32］　Huang L, Li S, Zhu A, et al. Noncontact body measurement for qinchuan cattle with LiDAR sensor ［J］. Sensors, 2018, 18(9): 3014.
［33］　Pezzuolo A, Guarino M, Sartori L, et al. Onbarn pig weight estimation based on body measurements by a Kinect v1 depth camera ［J］. Computers and Electronics in Agriculture, 2018, 148: 29-36.
［34］　Shi S, Yin L, Siang S, et al. Research on 3D surface reconstruction and body size measurement of pigs based on multiview RGBD cameras ［J］. Computers and Electronics in Agriculture, 2020, 175(1): 105543.
［35］　Shu Q, He X, Wang C, et al. Fast point cloud registration in multidirectional affine transformation ［J］. Optik, 2021, 229: 165884.
［36］　Lunin A, Glock C H. Systematic review of Kinectbased solutions for physical risk assessment in manual materials handling in industrial and laboratory environments ［J］. Computers & Industrial Engineering, 2021, 162: 107660.
［37］　郑立华, 王露寒, 王敏娟, 等. 基于Kinect相机的油麦菜自动化三维点云重建［J］. 农业机械学报, 2021, 52(7): 159-168.
Zheng Lihua, Wang Luhan, Wang Minjuan, et al. Automated 3D reconstruction of leaf lettuce based on Kinect camera ［J］.Transactions of the Chinese Society for Agricultural Machinery, 2021, 52(7): 159-168.
［38］　王庆闪, 张军, 刘元盛, 等. 基于NDT与ICP结合的点云配准算法［J］. 计算机工程与应用, 2020, 56(7): 88-95.Wang Qingshan, Zhang Jun, Liu Yuansheng, et al. Point cloud registration algorithm based on combination of NDT and ICP ［J］. Computer Engineering and Applications, 2020, 56(7): 88-95.
［39］　Aiger D, Mitra N J, CohenOr D. 4points congruent sets for robust pairwise surface registration ［J］. Transactions on Graphics, 2008, 27(3): 670-679.

[1]	韩梦杰, 刘发英, 杨振宇, , 孙卫孝, 陈肖, 魏忠彩, 李学强. 基于激光三维重建的种薯芽眼识别方法研究[J]. 中国农机化学报, 2024, 45(2): 200-206.
[2]	舒卓, 陈立平, 陈梅香, 张瑞瑞, 郭新宇, 温维亮. 单目多视角图像的美国白蛾三维模型重建系统与试验[J]. 中国农机化学报, 2023, 44(8): 168-173.
[3]	牟向伟, 孙国奇, 陈林涛, 于新业, 彭柱菁, 伍倩. 基于Kinect相机的多视角RGB-D信息融合的甜椒姿态估计研究[J]. 中国农机化学报, 2023, 44(10): 159-167.
[4]	钱美, 马保建, 李学志, 蒋焕煜. 基于Kinect V3的休眠期核桃树三维重建与试验[J]. 中国农机化学报, 2023, 44(10): 168-173.
[5]	耿丽杰, 顾健, 别晓婷, 冉维旭, 兰玉彬, . 基于Scan Context与NDT-ICP相融合的果园建图方法研究[J]. 中国农机化学报, 2022, 43(7): 44-50.
[6]	彭孝东, 时磊, 何静, 漆海霞, 兰玉彬, . 消费级RGB-D相机在农业领域应用现状与发展趋势[J]. 中国农机化学报, 2022, 43(4): 206-215.
[7]	林榆森, 李秋洁, 陈婷, . 基于快速点特征直方图的树木点云配准[J]. 中国农机化学报, 2022, 43(2): 155-162.
[8]	曹丽桃, 程曼, 袁洪波, 刘月琴, 随海燕, 赵晓霞. 可穿戴设备部署位置对羊只行为识别的影响与分析[J]. 中国农机化学报, 2022, 43(12): 133-141.
[9]	殷悦;张慧春;郑加强;. 基于双目立体视觉的植物三维重建系统[J]. 中国农机化学报, 2021, 42(3): 129-135.