A Study on the 3D Reconstruction Strategy of a Sheep Body Based on a Kinect v2 Depth Camera Array

Simple Summary Traditional methods for measuring animal body size primarily rely on manual data collection, which is not only costly and inefficient but also prone to significant data errors and can cause stress to the animals. To address these issues, this study introduces a 3D point cloud reconstruction method for sheep body measurement, achieving the synchronous collection and reconstruction of sheep body point cloud data. For the first time, a conditional voxel filtering box is proposed for downsampling, which effectively reduces the number of point clouds. A rotation normalization algorithm is used to correct the three views. Finally, a comparison is made between manual measurement speed, system measurement speed, and normalized system measurement speed. The results show that the normalized processing group has the most ideal time, with an average of 0.74 min to measure six indicators for one sheep. Lastly, the accuracy of the 3D point cloud reconstruction of the sheep body is verified, with a reconstruction accuracy error of 0.79% for body length as an example, indicating that the non-contact body measurement method based on 3D point cloud reconstruction is feasible and can provide an important reference for the breeding of superior breeds.Abstract Non-contact measurement based on the 3D reconstruction of sheep bodies can alleviate the stress response in sheep during manual measurement of body dimensions. However, data collection is easily affected by environmental factors and noise, which is not conducive to practical production needs. To address this issue, this study proposes a non-contact data acquisition system and a 3D point cloud reconstruction method for sheep bodies. The collected sheep body data can provide reference data for sheep breeding and fattening. The acquisition system consists of a Kinect v2 depth camera group, a sheep passage, and a restraining pen, synchronously collecting data from three perspectives. The 3D point cloud reconstruction method for sheep bodies is implemented based on C++ language and the Point Cloud Library (PCL). It processes noise through pass-through filtering, statistical filtering, and random sample consensus (RANSAC). A conditional voxel filtering box is proposed to downsample and simplify the point cloud data. Combined with the RANSAC and Iterative Closest Point (ICP) algorithms, coarse and fine registration are performed to improve registration accuracy and robustness, achieving 3D reconstruction of sheep bodies. In the base, 135 sets of point cloud data were collected from 20 sheep. After 3D reconstruction, the reconstruction error of body length compared to the actual values was 0.79%, indicating that this method can provide reliable reference data for 3D point cloud reconstruction research of sheep bodies.