Elliptical frustum-based volume estimation of strawberries from partial RGB-D point clouds using multi-view averaging

Reliable fruit volume estimation is essential for autonomous harvesting and precision yield management, but it remains challenging for strawberries due to their irregular shapes and the partial visibility inherent in single-view sensors. In this work, we present a geometric method that partitions segmented strawberry point clouds into horizontal cross-sections, fits an ellipse to each section, and aggregates elliptical frustums. To stabilize fitting in the presence of partial-view depth ambiguities, we introduce a height-dependent depth bias tailored to strawberry geometry. Volumes estimated independently from three viewpoints (front, left, right) are averaged to reduce viewpoint-dependent bias. Evaluation follows a two-stage design: (1) watertight strawberry meshes isolate geometric estimator performance (MAPE = 3.86 %, R 2 = 0.97), and (2) RGB-D acquisitions of 60 3D-printed surrogates assess the full pipeline under realistic noise. Multi-view averaging reduces MAPE from 10.40 % to 8.67 % (16.6 % relative improvement) and increases R 2 from 0.86 to 0.90 compared to single-view estimation. These results demonstrate that combining complementary viewpoints substantially improves robustness, making the approach viable for practical strawberry sizing applications.

Authors

Institutions

Publication Details

Journal
Frontiers in Plant Science
Published
2026-11-16
DOI
https://doi.org/10.3389/fpls.2026.1964134
Primary Topic
Smart Agriculture and AI
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Elliptical frustum-based volume estimation of strawberries from partial RGB-D point clouds using multi-view averaging

Marco Arts, Julius Diel, Mohieddine Jelali, Julia Brüninghaus et al.
6 links
Frontiers in Plant Science
Smart Agriculture and AI
article

Elliptical frustum-based volume estimation of strawberries from partial RGB-D point clouds using multi-view averaging

Marco Arts, Julius Diel, Mohieddine Jelali, Julia Brüninghaus, Julian Schrenk, Jonas Swiatek, Marlon Weiler
article en

Abstract

Reliable fruit volume estimation is essential for autonomous harvesting and precision yield management, but it remains challenging for strawberries due to their irregular shapes and the partial visibility inherent in single-view sensors. In this work, we present a geometric method that partitions segmented strawberry point clouds into horizontal cross-sections, fits an ellipse to each section, and aggregates elliptical frustums. To stabilize fitting in the presence of partial-view depth ambiguities, we introduce a height-dependent depth bias tailored to strawberry geometry. Volumes estimated independently from three viewpoints (front, left, right) are averaged to reduce viewpoint-dependent bias. Evaluation follows a two-stage design: (1) watertight strawberry meshes isolate geometric estimator performance (MAPE = 3.86 %, R 2 = 0.97), and (2) RGB-D acquisitions of 60 3D-printed surrogates assess the full pipeline under realistic noise. Multi-view averaging reduces MAPE from 10.40 % to 8.67 % (16.6 % relative improvement) and increases R 2 from 0.86 to 0.90 compared to single-view estimation. These results demonstrate that combining complementary viewpoints substantially improves robustness, making the approach viable for practical strawberry sizing applications.

Frontiers in Plant ScienceVol. 17
TH Köln - University of Applied Sciences (DE)
Zero hunger
Smart Agriculture and AI
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon