Calculating Volume of Pig Point Cloud Based on Improved Poisson Reconstruction

Area of Science:

• Agricultural Engineering
• Computer Vision
• Animal Science

Background:

• Pig point cloud data offers potential for digital reconstruction and weight estimation.
• Irregular shapes and missing data in pig point clouds hinder accurate volume calculation.
• Novel phenotype analysis using body part proportions can aid selective breeding.

Purpose of the Study:

• To develop and validate an improved Poisson reconstruction algorithm for pig point clouds.
• To accurately calculate pig body volume and estimate body weight.
• To explore the utility of volume-derived phenotypes in pig breeding.

Main Methods:

• An improved Poisson reconstruction algorithm was applied to pig body point clouds.
• Standard shape point clouds, a rabbit model, a piglet model, and 479 pig datasets were used for validation.
• Volume calculation and body weight estimation were performed using reconstructed point clouds.

Main Results:

• The improved Poisson reconstruction yielded smoother, more continuous, and complete pig body models.
• The piglet model volume calculation showed a 4% relative error.
• Weight estimation from volume had an average absolute error of 2.664 kg and a 2.478% relative error.
• A high correlation coefficient of 0.95 was found between pig body volume and body weight.

Conclusions:

• The improved Poisson reconstruction algorithm is accurate and reliable for pig point cloud analysis.
• Pig body volume is a strong predictor of body weight, with high correlation.
• This method provides a valuable tool for livestock weight estimation and developing new phenotypes for breeding programs.