Greenhouse structural steel pipes are susceptible to internal and external corrosion under hot and humid service conditions, leading to cross-sectional loss and reduced structural strength. This study proposes a cold infrared thermography method for screening corroded pipes and identifies effective indicators for different cooling-affected regions. Experiments were performed using three non-corroded and five salt-spray-corroded SPVHS-grade specimens and three field-corroded specimens. After a cooling stimulus was applied using an ice pack, time-series mean surface-temperature data were acquired using an infrared camera. For the direct cooling region, the recovery phase was fitted with an exponential model to estimate the time constant (τ). For the indirect cooling region, located approximately 50 mm from the direct cooling region, the cooling phase was fitted with a linear model to obtain R² and the root mean square error (RMSE). Mean τ was consistently higher in the non-corroded group than in the combined corroded group across all four environmental conditions. The linearity indices in the indirect cooling region showed only exploratory, condition-dependent group differences. These findings suggest that the recovery time constant may support corrosion screening under controlled conditions.

Advances in micro-electromechanical systems (MEMS) technology have markedly improved inertial sensors, driving down cost, size, and power consumption while enhancing performance. The resulting growth in supply and demand has created tiered performance classes and price points, enabling broader adoption of inertial sensing in agriculture. In this study, we comprehensively analyzed cases of agricultural application of inertial measurement unit (IMU) or single inertial sensors by classifying them into machine and living objects. In machine targeting research, it has been used as a key element in position estimation and navigation systems for autonomous driving of mobile objects such as agricultural vehicles and robots to detect rapid attitude changes and improve the accuracy of position estimation and path tracking of the mobile object. Cases targeting living objects can be largely divided into livestock and fruit trees. In the case of livestock, wearable IMUs were attached to body parts and combined with various deep learning techniques to detect specific behaviors of livestock or classify behavioral patterns. In the case of fruit trees, a study on vibration harvesting was conducted to analyze the vibration characteristics of fruit trees by attaching acceleration sensors to tree branches or fruits and to optimize vibration parameters. The analysis focused on the types of sensors and data utilization techniques used in each research case, and based on this, we plan to explore ways to utilize IMUs in agriculture in the future.

Deep learning technologies have been increasingly applied to precision weed management, and You Only Look Once (YOLO) has become widely adopted owing to its real-time detection capability and high accuracy. This systematic review analyzes YOLO-based weed recognition models with emphasis on field challenges and architectural optimization strategies. A structured search of Web of Science and Scopus, supplemented by manual searching, was conducted in accordance with the PRISMA 2020 guidelines. Following screening and eligibility assessment, 45 studies were included in the systematic synthesis. The main categories of analysis included attention mechanisms that enhance discrimination among vegetation with similar shapes and colors, multi-scale feature-fusion architectures that integrate information across feature pyramid levels to improve the detection of weeds of varying sizes and viewpoints, and lightweight architectures designed to support real-time operation on resource-constrained hardware, along with other optimization strategies. Recent developments increasingly combine these approaches to address crop–weed similarity, occlusion, scale variation, and complex field backgrounds while improving computational efficiency. The review further examines how architectural enhancement strategies contribute to detection accuracy and computational efficiency and discusses their implications for practical applications such as selective spraying and robotic weeding. Overall, this review links major field-level weed-recognition challenges with corresponding architectural optimization strategies and provides a systematic framework for developing accurate and efficient YOLO-based weed recognition systems for site-specific weed management.

Plastic greenhouses, which account for the majority of protected horticulture facilities in East Asia, are highly susceptible to wind-induced uplift failures that can lead to severe structural and economic damage. To address this issue, this study developed a low-power and low-cost wireless monitoring system applying the concept of structural health monitoring (SHM) to greenhouse foundations. Each sensor node integrates a MEMS-based inertial measurement unit (IMU) for attitude estimation, a LoRa module for long-range alert transmission, and a microSD module for data logging, while a gateway relays anomaly alerts to users through an IP network. Uplift tests were conducted on standard steel-pipe foundations commonly used in plastic greenhouses, and the proposed sensor nodes were evaluated alongside a commercial IMU to validate attitude estimation accuracy and anomaly detection performance. Despite the approximately 30-fold cost difference, comparable attitude estimation results were achieved. The system demonstrated low power consumption, confirming its feasibility for long-term operation using batteries or small solar cells. These results demonstrate the applicability of low-cost IMUs for real-time structural monitoring of lightweight greenhouse foundations.