Papers for
agriculture technology companies
Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.
Embedded system automates irrigation using soil sensors and weather forecasts
An Embedded Industrial Irrigation System with Weather-Integrated Intelligent Scheduling
Abstract: Efficient irrigation management requires timely responses to soil moisture conditions and anticipated weather changes. This article presents an embedded, weather-based irrigation system that integrates soil moisture monitoring using multiple sensors, edge-based decision-making, automated pump control, and local data management. Four capacitive soil moisture sensors are connected to an STM32F103C8T6 microcontroller for data acquisition, while a BTT Pi V1.2 platform evaluates the sensor data, retrieves short-term weather forecasts, makes irrigation decisions, manages a graphical user interface, and stores operational data locally. Individual sensor readings serve as the basis for zone-level irrigation decisions, while the median of four time-averaged readings supports monitoring and analysis. Forecasted precipitation is taken into account as effective precipitation to reduce the calculated irrigation requirement before the pump is activated. The system was evaluated as part of a one-month cultivation trial with radishes, during which a total of 38,112 data points from measurement cycles were recorded. The median value of normalized soil moisture averaged 47.93\% and remained predominantly within the configured control range. Three irrigation events with a planned total irrigation volume of 26.68 L were recorded. Finally, we demonstrate the technical feasibility of integrating sensors, weather data, local data processing, automated control, and data collection into a scalable platform for precision irrigation.
Multimodal continual learning improves leaf counting accuracy
MCLC-NET: Multimodal Continual Learning for Leaf Counting
Abstract: Leaf counting is an important task in plant phenotyping for monitoring plant growth and estimating crop yield. Most existing methods rely on RGB images, but their performance is often affected by occlusion, lighting variations, and other real-world challenges. Additional modalities, such as depth and thermal images, can provide useful complementary information. However, multimodal leaf counting remains underexplored. Also, many existing methods assume that all training data are available simultaneously, which is impractical in real agricultural settings, where data is collected over time from multiple sources. To address these challenges, we propose MCLC-NET, a multimodal continual learning framework for leaf counting. It learns tasks sequentially using a memory-based strategy with a memory buffer to retain important samples from previous tasks. We also introduce MMLC, a real-world multimodal leaf-counting dataset designed for a domain incremental scenario (DIS) in CL. It contains RGB, depth, and thermal images collected across different crop types under varying environmental conditions, arranged in three orderings: crop-wise, time-wise, and mixed. Experimental results, averaged over three random seeds, demonstrate that MCLC-NET consistently outperforms existing methods across all three task orderings, achieving the lowest AMSE of 0.675$\pm$0.027, 0.542$\pm$0.069, and 0.745$\pm$0.057, respectively.