Search papers, labs, and topics across Lattice.
This study introduces a novel annotated image time-series dataset of 691 sweet pepper plants, featuring 4,837 images and per-plant fruit counts categorized by maturity, addressing the scarcity of public datasets for precision agriculture. A multimodal deep learning framework is proposed that integrates high-dimensional image features from the DinoV3 encoder with numerical count measurements, utilizing an LSTM network to model temporal dependencies and manage irregular sampling intervals. The results show a significant reduction in RMSE by 33% and 38% compared to a persistence baseline in the 2022 and 2023 seasons, respectively, while also providing calibrated uncertainty estimates that enhance decision-making in agricultural practices.
A multimodal deep learning framework improves yield forecasting accuracy for sweet peppers by over 30%, setting a new standard for precision agriculture.
Accurate yield forecasting at the individual-plant level is critical for precision agriculture and supply-chain planning, yet public datasets capturing both visual growth dynamics and per-plant measurement labels are scarce. In this paper, we introduce a novel, annotated image time-series dataset of 691 sweet pepper plants monitored over two growing seasons, comprising 4837 images with per-plant fruit counts categorized by maturity. We propose a multimodal deep learning framework that fuses high-dimensional image features, extracted using the DinoV3 encoder, with numerical count measurements. Our architecture utilizes a Long Short-Term Memory (LSTM) network to model temporal dependencies and handles irregular sampling intervals common in greenhouse monitoring. Through quantitative experiments, we demonstrate that this multimodal approach reduces RMSE over a persistence baseline by 33% and 38% in the 2022 and 2023 seasons, respectively, with a further 1.2% average gain over a measurement-only model. Furthermore, we employ Deep Ensembles and Gaussian Negative Log-Likelihood (NLL) to provide calibrated uncertainty estimates, with an Uncertainty Calibration Error (UCE) ranging from 0.39 to 0.89 depending on the cross-season evaluation direction, offering a principled confidence signal for real-world agricultural decision-making. We release the dataset and code to support reproducible research and to accelerate development of data-driven yield forecasting methods for horticultural crops.
