Vision Based Automated Point-Cloud Processing Pipeline For High Throughput Phenotyping

High throughput phenotyping is rapidly gaining widespread popularity due to its non-destructive approach for plant traits extraction. In this study, we focus on developing a vision based automated 3D point cloud processing pipeline for accurate estimation of plant traits, namely - plant height, leaf area index(LAI), and leaf inclination. Furthermore, the obtained estimates are validated by comparing the results with LeasyScan data in terms of coefficient of determination (R2), root mean squared error (RMSE), and correlation coefficient. These metrics are found to be around 0.90, 0.10, and 0.96, respectively, for each of the traits. Regression analysis has also been performed to gain some analytical insights on the data