Plant diseases cause significant damage to agriculture, leading to substantial yield losses and posing a major threat to food security. Detection, identification, quantification, and diagnosis of plant diseases are crucial parts of precision agriculture and crop protection. Modernizing agriculture and improving production efficiency are significantly affected by using computer vision technology for crop disease diagnosis. This technology is notable for its non-destructive nature, speed, real-time responsiveness, and precision. Deep learning (DL), a recent breakthrough in computer vision, has become a focal point in agricultural plant protection that can minimize the biases of manually selecting disease spot features. This study reviews the techniques and tools used for automatic disease identification, state-of-the-art DL models, and recent trends in DL-based image analysis. The techniques, performance, benefits, drawbacks, underlying frameworks, and reference datasets of more than 278 research articles were analyzed and subsequently highlighted in accordance with the architecture of computer vision and deep learning models. Key findings include the effectiveness of imaging techniques and sensors like RGB, multispectral, and hyperspectral cameras for early disease detection. Researchers also evaluated various DL architectures, such as convolutional neural networks, vision transformers, generative adversarial networks, vision language models, and foundation models. Moreover, the study connects academic research with practical agricultural applications, providing guidance on the suitability of these models for production environments. This comprehensive review offers valuable insights into the current state and future directions of deep learning in plant disease detection, making it a significant resource for researchers, academicians, and practitioners in precision agriculture.

Land-use/land-cover (LULC) change is considered a key human factor influencing groundwater recharge in floodplains. Without accurate estimations, the impact of LULC change on water balance components may be either significantly understated or exaggerated. This paper assesses the impacts of LULC changes from 1990 to 2018 on water balance components and groundwater levels of the Drava floodplain, Hungary, where human interference has led to a critical environmental situation. In this study, a spatially-distributed water balance model (WetSpass-M), and a groundwater flow model (MODFLOW-NWT) were integrated to assess the impacts of LULC changes. The moderate expansion of built-up areas increased surface runoff, while the afforestation of arable land and meadows and the overgrowth of bare mudflats with willow shrubs increased evapotranspiration. As a consequence, total annual groundwater recharge decreased by 5.3 × 10⁷ m³ in the floodplain with an average of 335 mm year⁻¹ and 317 mm year⁻¹ in 2012 and 2018, respectively. Moreover, an average groundwater level decline by 0.1 m is observed in the same period. Declined groundwater recharge, increased runoff, and evapotranspiration exerted a negative effect on water resources in the Drava basin. The approach tested in this paper allows temporal and spatial estimation of hydrological components under the changes of LULC, providing quantitative information for decision-makers and stakeholders to implement efficient and sustainable management of water resources in the Drava floodplain. The provided integrated model is also applicable to regionally.

The assessment of spatial and temporal distribution of groundwater recharge is required as an input to develop the regional groundwater model in the Drava flood plain for more accurate simulations of different management scenarios. WetSpass-M, a GIS-based spatially-distributed water balance model, was implemented to assess monthly, seasonal, and the annual averages of groundwater recharge, surface runoff and actual evapotranspiration in the Drava basin, Hungary for the period between 2000–2018. The basic relevant input-data for the Wetspass-M model is prepared in grid-maps using the tool ARCGIS tool. It comprises monthly climatological recordings (e.g., rainfall, temperature, wind speed), distributed land cover, soil map, groundwater depth, topography, and slope. The long-term temporal and spatial average monthly precipitation (58 mm) is distributed as 29% (17 mm) surface runoff, 27% (16 mm) actual evapotranspiration, and 44% (25 mm) groundwater recharge. The mean annual groundwater recharge, actual evapotranspiration, and surface runoff were 307, 190, and 199 mm, respectively. The findings of the WetSpass-M model are intended to support integrated groundwater modeling. The analysis of simulation results shows that WetSpass-M model works properly to simulate hydrological water budget components in the Drava basin. Moreover, a better understanding of the simulated long-term average spatial distribution about water balance components is useful for managing and planning the available water resources in the Drava basin.

Urban surfaces are often covered by impermeable materials such as concrete and asphalt which intensify urban runoff and pollutant concentration during storm events, and lead to the deterioration of the quality of surrounding water bodies. Detention ponds are used in urban stormwater management, providing two-fold benefits: flood risk reduction and pollution load minimization. This paper investigates the performance of nine proposed detention ponds (across the city of Renton, Washington, USA) under different climate change scenarios. First, a statistical model was developed to estimate the pollutant load for the current and future periods and to understand the effects of increased rainfall on stormwater runoff and pollutant loads. The Personal Computer Storm Water Management Model (PCSWMM) platform is employed to calibrate an urban drainage model for quantifying stormwater runoff and corresponding pollutant loads. The calibrated model was used to investigate the performance of the proposed nine (9) detention ponds under future climate scenarios of 100-year design storms, leading to identifying if they are likely to reduce stormwater discharge and pollutant loads. Results indicated significant increases in stormwater pollutants due to increases in rainfall from 2023 to 2050 compared to the historical period 2000–2014. We found that the performance of the proposed detention ponds in reducing stormwater pollutants varied depending on the size and location of the detention ponds. Simulations for the future indicated that the selected detention ponds are likely to reduce the concentrations (loads) of different water quality constituents such as ammonia (NH₃), nitrogen dioxide (NO₂), nitrate (NO₃), total phosphate (TP), and suspended solids (SS) ranging from 18 to 86%, 35–70%, 36–65%, 26–91%, and 34–81%, respectively. The study concluded that detention ponds can be used as a reliable solution for reducing stormwater flows and pollutant loads under a warmer future climate and an effective adaptation option to combat climate change related challenges in urban stormwater management.

The experiments were conducted at different levels of infrared power, airflow, and temperature. The relationships between the input process factors and response factors’ physicochemical properties of dried garlic were optimized by a self-organizing map (SOM), and the model was developed using machine learning. Artificial Neural Network (ANN) with 99% predicting accuracy and Self-Organizing Maps (SOM) with 97% clustering accuracy were used to determine the quality characteristics of garlic. Specifically, five key areas were identified, and valuable insights were offered for optimizing garlic production and improving its overall quality. The (aw) values for the sample ranged from 0.43 to 0.48. The maximum vitamin C content was 0.112 mg/g, followed by an air temperature of 40 °C and 0.7 m/s air velocity under 1500 W/m². The total color change values increased with IR and higher air temperature but declined with higher air velocity. Also, the garlic’s flavor strength, allicin content, water activity, and vitamin C levels decreased as the IR and air temperature increased. The results demonstrated a significant impact of the independent parameters on the response parameters (P < 0.01). Interestingly, the machine learning predictions closely matched the test data sets, providing valuable insights for understanding and controlling the factors affecting garlic drying performances.