3D Reconstruction-Based Seed Counting of Sorghum Panicles for Agricultural Inspection

TL;DR

This paper introduces a novel 3D reconstruction method for sorghum panicles that uses seed landmarks to improve seed counting accuracy, outperforming traditional 2D image extrapolation methods.

Contribution

The paper presents a new 3D reconstruction approach with a seed-based semantic landmark system and a ground-truth-free quality metric for phenotyping in sorghum breeding.

Findings

Enhanced seed count accuracy over 2D extrapolation methods

Effective 3D modeling of sorghum panicles for phenotyping

New metric for assessing point cloud quality without ground truth

Abstract

In this paper, we present a method for creating high-quality 3D models of sorghum panicles for phenotyping in breeding experiments. This is achieved with a novel reconstruction approach that uses seeds as semantic landmarks in both 2D and 3D. To evaluate the performance, we develop a new metric for assessing the quality of reconstructed point clouds without having a ground-truth point cloud. Finally, a counting method is presented where the density of seed centers in the 3D model allows 2D counts from multiple views to be effectively combined into a whole-panicle count. We demonstrate that using this method to estimate seed count and weight for sorghum outperforms count extrapolation from 2D images, an approach used in most state of the art methods for seeds and grains of comparable size.