AI Science Tracker
EN

Autonomous apple fruitlet sizing and growth rate tracking using computer vision

Not stated in the reviewed arXiv v1 · 3 Dec 2022 · other

Discovery

Autonomous apple fruitlet sizing and growth rate tracking using computer vision

Source check: Source checked. Not independent scientific confirmation.

Finding: Computer-vision models segment and associate apple fruitlets across days to estimate growth and thinning response. The first arXiv version reports orchard and robotic-field evaluations, with predicted persist/abscise rates within 3% of manual-caliper estimates; this is a measurement result, not a demonstrated yield gain.

Technology: Mask R-CNN and pix2pix fruitlet segmentation with an attentional graph neural network for cross-day association

Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.

AI contribution
Mask R-CNN and pix2pix detect and segment fruitlets; an attentional graph neural network associates individuals across days for growth/thinning estimates.
Test and result
The v1 pipeline estimated fruitlet persistence/abscission rates within 3% of manual-caliper estimates and evaluated orchard growth tracking and a robotic field image set.
Limitation / still unknown
Results are limited to evaluated sites and varieties; performance does not demonstrate increased yield or downstream agronomic benefit.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://arxiv.org/html/2212.01506v1

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.

Review date: 22 Nov 2026. Review scope: Primary arXiv v1 full text, Methods and Results; version-specific figures and chronology checked..

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 3/5 · Validation 3/5 · Novelty 4/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open in the full catalog for filters, charts and related discoveries. Browse other discoveries.