Beetroot type classification based on 3D structure constructed with neural radiance fields

doi:10.1016/j.compag.2026.111790

Beetroot type classification based on 3D structure constructed with neural radiance fields

2026-04-20

0

PRE

AI

DOI：10.1016/j.compag.2026.111790

Original

Original request for help

Save

Abstract

En 中文

• NeRF achieved high-fidelity 3D reconstruction of beetroots with fewer artifacts than SfM-MVS. • An end-to-end network, Point-SugarRoot, was proposed by integrating KAN and CBAM modules. • Point-SugarRoot surpassed PointNet++ and outperformed 2D image and phenotype paths. • This research offered valuable technical insights for cultivar breeding and crop management.

Keywords:

Neural Radiance Fields

Beetroot Classification

3D Reconstruction

Point-SugarRoot

Cultivar Breeding

Beetroot type classification based on 3D structure constructed with neural radiance fields

Abstract

Journal

Organization