![[2503.22809] Data-Driven Worker Activity Recognition and Efficiency Estimation in Manual Fruit Harvesting](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2503.22809] Data-Driven Worker Activity Recognition and Efficiency Estimation in Manual Fruit Harvesting
Summary
This study presents a data-driven system for recognizing worker activities and estimating efficiency in manual fruit harvesting, specifically focusing on strawberry picking.
Why It Matters
Understanding worker efficiency in agriculture is crucial for optimizing labor management and improving productivity. This research provides a technological solution that can significantly reduce non-productive time during harvesting, which is vital for the agricultural sector facing labor shortages and increasing demand.
Key Takeaways
- Developed a system using instrumented carts to track picker activities.
- Achieved a high activity recognition performance with an F1 score of 0.97.
- Estimated average picker efficiency at 75.07% with high accuracy.
- The technology can help optimize harvest processes and reduce inefficiencies.
- Potential for broader applications in agricultural labor management.
Computer Science > Machine Learning arXiv:2503.22809 (cs) [Submitted on 28 Mar 2025 (v1), last revised 13 Feb 2026 (this version, v3)] Title:Data-Driven Worker Activity Recognition and Efficiency Estimation in Manual Fruit Harvesting Authors:Uddhav Bhattarai, Rajkishan Arikapudi, Steven A. Fennimore, Frank N Martin, Stavros G. Vougioukas View a PDF of the paper titled Data-Driven Worker Activity Recognition and Efficiency Estimation in Manual Fruit Harvesting, by Uddhav Bhattarai and 4 other authors View PDF HTML (experimental) Abstract:Manual fruit harvesting is common in agriculture, but the amount of time pickers spend on non-productive activities can make it very inefficient. Accurately identifying picking vs. non-picking activity is crucial for estimating picker efficiency and optimising labour management and harvest processes. In this study, a practical system was developed to calculate the efficiency of pickers in commercial strawberry harvesting. Instrumented picking carts (iCarritos) were developed to record the harvested fruit weight, geolocation, and iCarrito movement in real time. The iCarritos were deployed during the commercial strawberry harvest season in Santa Maria, CA. The collected data was then used to train a CNN-LSTM-based deep neural network to classify a picker's activity into "Pick" and "NoPick" classes. Experimental evaluations showed that the CNN-LSTM model showed promising activity recognition performance with an F1 score of 0.97. The recognitio...
