What does this research mean for the field?

Video-based multi-object tracking systems can effectively automate apple orchard fruit load estimation, achieving the highest accuracy when utilizing longer scanning distances, bilateral tree scanning, and timing near harvest. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To evaluate a video multi-object tracking system for automated fruit load estimation across varying scanning distances, sensor types, and dates in an apple orchard.

September 5, 2026Computers and Electronics in AgricultureOpen Access

Video-based fruit detection and tracking: effects of scanning conditions on fruit load estimation

View Full Paper

Ask AI

Bookmark

To evaluate a video multi-object tracking system for automated fruit load estimation across varying scanning distances, sensor types, and dates in an apple orchard.
Tested a computer vision pipeline integrating fruit detection, multi-object tracking, orchard localization, and yield map generation across 420 apple trees.
Captured video data using two RGB-D cameras (Azure Kinect DK and ZED 2) at three scanning distances (125 cm, 175 cm, and 225 cm) across two pre-harvest dates, comparing unilateral versus bilateral canopy scanning.
The Azure Kinect provided more consistent performance across dates, while a longer scanning distance of 225 cm improved accuracy over close-up partial views by capturing the entire tree.
The highest accuracy occurred near harvest with the ZED 2 sensor at a 225 cm distance, achieving a Mean Absolute Percentage Error (MAPE) of 6.91% and a coefficient of determination (R²) of 0.733.
Bilateral row scanning outperformed single-sided scanning by capturing fruit counts across both sides of the canopy tree stretch.

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Experimental study demonstrates video tracking improves apple load estimation under varying scanning distances and sensor conditions, highlighting optimal setups for automated orchard yield mapping.

Cite This Study

Felip-Pomés et al. (2026) studied this question.

synapsesocial.com/papers/6a9bd4046b95aff0620eb47d https://doi.org/10.1016/j.compag.2026.112330

View Full Paper

Ask AI

Bookmark

Implication

Experimental study demonstrates video tracking improves apple load estimation under varying scanning distances and sensor conditions, highlighting optimal setups for automated orchard yield mapping.

Video-based fruit detection and tracking: effects of scanning conditions on fruit load estimation

Key Points

Discussion

Implication

Cite This Study

Video-based fruit detection and tracking: effects of scanning conditions on fruit load estimation

Key Points

Discussion

Implication

Cite This Study

Also Consider

Also Consider