Oluwasegun Moses Ogundele

Reliable individual tracking supports animal behavior monitoring in precision livestock farming. However, tracking cattle using a single camera in high-density barns is challenging because of frequent occlusion, scale variation, and irregular movement. This study presents a lightweight multi-object tracking framework for densely housed Hanwoo cattle, combining a Hanwoo-trained YOLOv8n detector with appearance-free BoT-SORT association. Frame sampling reduces the detection workload, while an offline Multi-Temporal Window Repair (MTW-Repair) procedure reconnects fragmented trajectories. The framework was evaluated using short annotated video clips and a continuous 215 min barn recording. In the short clips, appearance-free BoT-SORT outperformed ByteTrack while maintaining comparable throughput, and MTW-Repair reduced identity switches across all evaluated frame-sampling intervals. In the long recording, MTW-Repair reduced the total number of generated identities for both trackers. Manual verification of 50 randomly sampled ByteTrack merge events yielded a merge precision of 84.0%. Appearance-based re-identification offered only marginal accuracy gains at a higher computational cost. These findings support the potential of frame sampling and offline trajectory repair for computationally efficient cattle monitoring.

Increasing global temperatures and unpredictable weather threaten agricultural production, intensifying the reliance on greenhouse cultivation. While naturally ventilated greenhouses offer an energy-efficient solution, their thermal performance under high heat loads is a growing concern. This study developed and validated a 3D CFD model to investigate the temperature distribution inside a single-span, naturally ventilated greenhouse. The model, implemented in ANSYS Fluent, considered the influence of external meteorological conditions, including solar radiation and wind, on the internal microclimate. The model demonstrated high accuracy, with an RMSE of 0.847°C and an MAE of 0.714°C for temperature when compared to experimental data. Diurnal analysis confirmed the dominant role of solar radiation, showing a strong correlation between indoor temperature and external solar radiation (r = 0.96) on sunny days. Spatially, the simulation revealed extreme thermal heterogeneity under high solar load, with a total internal temperature variation of up to 25°C (from 296 K to 321 K). The distribution was characterized by a distinct vertical stratification due to buoyancy and a strong longitudinal gradient, with the hottest air accumulating in the upper leeward corner. These findings highlight areas of potential crop heat stress and demonstrate the limitations of side-ventilation alone. The validated model serves as a valuable tool for optimizing ventilation strategies and improving climate control system design.

Vertical farming or vertical growing systems, a subset of controlled environment agriculture (CEA), integrates smart and precision agriculture to optimize crop production in stacked environments within a greenhouse, utilizing Internet of Things (IoT) sensors, automation, and advanced lighting. However, vertical farming can be carried out in the open, but the advantages of a closed system are sacrificed. This review examines recent advancements, including precision irrigation (hydroponics, aeroponics, and aquaponics), LED spectral optimization, robotics, and machine learning (ML) for enhanced resource efficiency. Case studies from Singapore, Japan, and the Netherlands demonstrate yield increases of 200-300% and energy savings of 30-50% compared to traditional methods. By integrating robotics, ML, and IoT, yields can be maximized, resource utilization can be optimized, and environmental impact can be reduced. Case studies from Singapore, Japan, and the Netherlands demonstrate the potential, with kale yields of 300 t•ha-1 , strawberry production at 200 t•ha-1 , and dual basil fish outputs achieved through 98% water recycling. Challenges such as high costs, energy demands, and technical complexity persist, particularly in developing regions. Future directions propose the integration of renewable energy, can offset energy reliance, cutting costs by 50%, low-cost sensors that broaden access and simplify IoT deployment, and modular designs that enhance scalability and offer hope for adaptation. This manuscript synthesizes quantitative metrics, comparative analyses, and implementation challenges, providing engineers and researchers with a detailed framework to advance sustainable urban agriculture through technology-driven solutions.

Growing interest in long-duration space missions and extraterrestrial colonization has driven the development of reliable space agriculture systems essential for sustaining life during these missions. However, the effects of environmental stressors such as low atmospheric pressure on plant growth have not yet been fully elucidated, which is crucial for the design of bioregenerative life support systems. This study investigated the growth responses of kale (Brassica oleracea var. acephala) under different pressure conditions (101.3, 67, and 33.0 kPa) in a closed environment, simulating potential extraterrestrial habitats. Two weeks after germination, kale plants were cultivated in custom-made acrylic pressure chambers for 19 days. Water (20 mL) was supplied on days 1 and 9. Leaf length, leaf width, and chlorophyll content were measured on days 1, 9, and 19. No significant differences were observed among the groups on day 1 (p > 0.05). However, by day 9, significant increases in leaf growth were observed in groups A (101.3 kPa) and B (67 kPa), with group B showing temporary enhancement of leaf growth (p < 0.05). By day 19, group A exhibited the highest growth rates for all parameters (leaf length 44.72 ± 2.74 mm, leaf width 39.19 ± 5.99 mm), while group B maintained relatively high growth, and group C (33 kPa) displayed the lowest growth rates, indicating a tendency for growth inhibition under prolonged hypobaric conditions. Chlorophyll content (SPAD values) increased during the early stages, with the highest values recorded in group C, suggesting that low pressure may temporarily stimulate chlorophyll preservation and accumulation. However, by day 19, chlorophyll content decreased across all groups, indicating challenges in maintaining photosynthetic capacity over the long term. These findings suggest that while low-pressure environments may initially promote physiological activity, they ultimately lead to growth inhibition. This study highlights the importance of maintaining appropriate pressure levels for space-based crop cultivation and provides foundational data for designing bioregenerative life support systems.

Accurate fruit counting is crucial for data-driven decision-making in modern precision agriculture. In strawberry cultivation, a labor-intensive sector, automated, scalable yield estimation is especially critical. However, dense foliage, variable lighting, visual ambiguity of ripeness stages, and fruit clustering pose significant challenges. To overcome these, we developed a real-time multi-stage framework for strawberry detection and counting by optimizing a YOLOv8s detector and integrating a class-aware tracking system. The detector was enhanced with a lightweight C3x module, an additional detection head for small objects, and the Wise-IOU (WIoU) loss function, thereby improving performance against occlusion. Our final model achieved a 92.5% mAP@0.5, outperforming the baseline while reducing the number of parameters by 27.9%. This detector was integrated with the ByteTrack multiple object tracking (MOT) algorithm. Our system enabled accurate, automated fruit counting in complex greenhouse environments. When validated on video data, results showed a strong correlation with ground-truth counts (R² = 0.914) and a low mean absolute percentage error (MAPE) of 9.52%. Counting accuracy was highest for ripe strawberries (R² = 0.950), confirming the value for harvest-ready estimation. This work delivers an efficient, accurate, and resource-conscious solution for automated yield monitoring in commercial strawberry production.

This study quantitatively evaluated the disinfection performance of a low-pressure mercury UV-C system against the pathogenic bacterium Agrobacterium rhizogenes by analyzing the effect of different combinations of lamp power consumption and flow rate. Agricultural water samples were collected from three commercial farms (Sources A, B, and C) in Sacheon-si, Gyeongsangnam-do, for the first experiment, and Source D in Gunbuk-myeon, Haman-gun, Republic of Korea, for the second experiment. All water samples were artificially inoculated with A. rhizogenes prior to treatment. The experiment tested nine combinations of three lamp power consumption levels (40, 75, and 105 W) and three flow rates (10, 15, and 30 L/min). Among these, samples #10, #20, and #30 were inoculated but not UV-treated, serving as the corresponding pre-disinfection controls. In the second experiment, a fixed power of 75 W was applied under high-flow conditions (60 and 120 L/min), with samples #37 and #38 serving as the corresponding untreated controls. In all tested conditions, absence of detectable bacteria was confirmed, except for sample #33 (75 W, 60 L/min), in which viable cells were detected at 2.0×10 2 CFU/mL. However, this represented approximately a 4.9 log reduction compared to its control (#37, 1.6×10⁷ CFU/mL), indicating substantial bactericidal efficacy. These findings suggest that effective disinfection can be achieved even under high-flow conditions, provided that sufficient UV dose is delivered. The results underscore the practical applicability of low-pressure UV-C systems in agricultural water treatment through the optimal configuration of operating parameters.

Accurate localization of picking points and depth estimation is critical for implementing a robotic strawberry harvesting system. Due to the delicate nature of strawberries, harvesting must be performed without bruising or damage, typically by grasping and cutting the peduncle of the ripe strawberry. However, accurately detecting and localizing the thin peduncle in a cluttered environment is a significant challenge. This study proposed depth fused Mask R-CNN (DF-Mask R-CNN), which integrates depth information of the scene with the RGB image to enhance the detection, localization, and segmentation of strawberries and their peduncles in a greenhouse environment. To generate a dense depth map, a cutting-edge monocular depth estimator, ZoeDepth was used. The proposed DF-Mask R-CNN with ResNet101-FPN exhibited superior instance segmentation performance, with an overall mAP of 81.9%, with mAP_small at 33.3%, mAP_medium at 78.79%, mAP_large at 88.8 and AP^IOU=0.5 at 98.1%. In tests with 300 ripe strawberry samples, the method demonstrated a robust picking point detection, with a mean absolute error and root mean square error of 1.98 cm and 2.12 cm, respectively. These results highlight the effectiveness of the DF-Mask R-CNN model combined with the ZoeDepth estimator in enhancing the detection, localization, and segmentation of strawberries and their peduncles. This approach enables precise picking point localization and depth estimation for efficient vision systems for robotic strawberry harvesting.

Carbon dioxide (CO₂) concentrations play an important role in plant production, as they have a direct impact on both plant growth and yield. Therefore, the objectives of this study were to predict CO₂ concentrations in the greenhouse by applying time series models using five datasets. To estimate the CO₂ concentrations, this study was conducted over a four-month period from 1 December 2023 to 31 March 2024, in a strawberry-cultivating greenhouse. Fifteen sensors (MCH-383SD, Lutron, Taiwan) were installed inside the greenhouse to measure CO₂ concentration at 1-min intervals. Finally, the dataset was transformed into intervals of 1, 5, 10, 30, and 60 min. The time-series data were analyzed using the autoregressive integrated moving average (ARIMA) and the Prophet Forecasting Model (PFM), with performance assessed through root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R²). The evaluation indicated that the best model performance was achieved with data collected at 1-min intervals, while model performance declined with longer intervals, with the lowest performance observed at 60-min intervals. Specifically, the ARIMA model outperformed across all data collection intervals while comparing with the PFM. The ARIMA model, with data collected at 1-min intervals, achieved an R² of 0.928, RMSE of 7.359, and MAE of 2.832. However, both ARIMA and PFM exhibited poorer performances as the interval of data collection increased, with the lowest performance at 60-min intervals where ARIMA had an R² of 0.762, RMSE of 19.469, and MAE of 11.48. This research underscores the importance of frequent data collection for precise environmental control in greenhouse agriculture, emphasizing the critical role of short-interval data collection for accurate predictive modeling.