Laterite is the predominant zonal soil in China’s southernmost tropical rainforest and monsoon forest regions, where typhoons are the primary source of precipitation. These storms pose significant risks of land and soil degradation due to heavy rainfall. In recent years, a substantial area of sloping land has been converted to agricultural use in these regions, predominantly for the cultivation of crops grown in laterite soil. These activities contribute to soil erosion, exacerbate environmental challenges, and hinder the pursuit of sustainable development. There is a paucity of research reports on the processes and mechanisms of runoff and sediment on sugarcane-cropped slopes in regions with laterite soil under heavy rainfall conditions. In this study, four different heavy rainfall scenarios of 75, 100, 125, and 150 mm/h were designed to assess the impact on sugarcane growth at four key stages and to measure the resulting effects on initial runoff time, surface runoff, and sediment yield from laterite soil slopes under controlled laboratory conditions. The results showed that the Horton model explained much of the variation in infiltration rate on the sugarcane-cropped laterite slopes. The cumulative sediment yield on the sugarcane-cropped laterite slopes followed a second-degree polynomial function. The initial runoff time, infiltration intensity, runoff intensity, and sediment yield were all linearly related to the leaf area index (LAI) and rainfall intensity on the sugarcane-cropped slope surface. The leaf area index exerted a greater influence on the initial runoff time and infiltration intensity than rainfall intensity. However, rainfall intensity exerted a greater influence on the runoff intensity and sediment yield than the leaf area index. Compared with the bare sloping land, the average sediment yield was reduced by 12.2, 33.1, 58.2, and 64.9% with the sugarcane growth stages of seedling, tillering, elongation, and maturity, respectively.

The fast human climate change we are witnessing in the early twenty-first century is inextricably linked to the health and function of the biosphere. Climate change is affecting ecosystems through changes in mean conditions and variability, as well as other related changes such as increased ocean acidification and atmospheric CO 2 concentrations. It also interacts with other ecological stresses like as degradation, defaunation, and fragmentation.Ecology and climate monitoring are critical to understanding the complicated interactions between ecosystems and changing climate trends. This review paper dives into the issues of ecological and climate monitoring, emphasizing the complications caused by technical limits, data integration, scale differences, and the critical requirement for accurate and timely information. Understanding the ecological dynamics of these climatic impacts, identifying hotspots of susceptibility and resistance, and identifying management measures that may aid biosphere resilience to climate change are all necessary. At the same time, ecosystems can help with climate change mitigation and adaptation. The processes, possibilities, and constraints of such nature-based climate change solutions must be investigated and assessed. Addressing these issues is critical for developing successful policies and strategies for mitigating the effects of climate change and promoting sustainable ecosystem management. Human actions inscribe their stamp in the big narrative of our planet’s story, affecting the very substance of the global atmosphere. This transformation goes beyond chemistry, casting a spell on the physical characteristics that choreograph Earth’s brilliant dance. These qualities, like heavenly notes, create a song that echoes deep into the biosphere. We go on a journey via recorded tales of ecological transformation as they respond to the ever-shifting environment in this text. We peek into the rich fabric of change, drawing insight from interconnected observatories. Nonetheless, this growing symphony is set to unleash additional transformational stories - narratives of natural riches and rhythms that are both economically and environmentally essential. Understanding these stories is essential for navigating this developing epic. A roadmap for sustainable development necessitates the ability to comprehend these stories, a problem that resonates across the breadth of monitoring programs, particularly in the infancy of integrated sites.

This paper presents a novel approach to time series forecasting, an area of significant importance across diverse fields such as finance, meteorology, and industrial production. Time series data, characterized by its complexity involving trends, cyclicality, and random fluctuations, necessitates sophisticated methods for accurate forecasting. Traditional forecasting methods, while valuable, often struggle with the non-linear and non-stationary nature of time series data. To address this challenge, we propose an innovative model that combines signal decomposition and deep learning techniques. Our model employs Generalized Autoregressive Conditional Heteroskedasticity (GARCH) for learning the volatility in time series changes, followed by Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) for data decomposition, significantly simplifying data complexity. We then apply Graph Convolutional Networks (GCN) to effectively learn the features of the decomposed data. The integration of these advanced techniques enables our model to fully capture and analyze the intricate features of time series data at various interval lengths. We have evaluated our model on multiple typical time-series datasets, demonstrating its enhanced predictive accuracy and stability compared to traditional methods. This research not only contributes to the field of time series forecasting but also opens avenues for the application of hybrid models in big data analysis, particularly in understanding and predicting the evolution of complex systems.

With the continuous advancement of science and technology, there has been a growing awareness of safety among people worldwide. Natural disasters such as wildfires, earthquakes, and floods pose persistent threats to both lives and property on our planet, which serves as our fundamental habitat. While it is impossible to prevent or entirely avert these calamities, rapid identification of affected areas and prompt damage assessment post-disaster can significantly aid in the formulation of effective rescue strategies, ultimately saving more lives. This article delves into the application of transfer learning in satellite image damage assessment—a methodology that involves transferring previously acquired knowledge to enhance a model's adaptability to new tasks. Given the limited availability of datasets for satellite image analysis, transfer learning proves to be an effective approach. Specifically, the study proposes a transfer learning method based on YOLOv5 for satellite image damage assessment. Initially, a general convolutional neural network model is trained using a substantial dataset of natural images. Subsequently, the early layers of this model are frozen, while the later layers undergo training to adapt to satellite image data. Fine-tuning is then employed to further enhance the overall model performance. The results demonstrate that this approach yields a high accuracy rate in satellite image damage assessment. Moreover, compared to conventional deep learning methods, the proposed method effectively leverages pre-trained models' knowledge, thereby reducing data dependency. Additionally, it displays robust generalization capabilities across diverse tasks and datasets, underscoring its potential for facilitating transfer learning across various domains.

Tomato disease control remains a major challenge in the agriculture sector. Early stage recognition of these diseases is critical to reduce pesticide usage and mitigate economic losses. While many research works have been inspired by the success of deep learning in computer vision to improve the performance of recognition systems for crop diseases, few of these studies optimized the deep learning models to generalize their findings to practical use in the field. In this work, we proposed a model for identifying tomato leaf diseases based on both in-house data and public tomato leaf images databases. Three deep learning network architectures (VGG16, Inception_v3, and Resnet50) were trained and tested. We packaged the trained model into an Android application named TomatoGuard to identify nine kinds of tomato leaf diseases and healthy tomato leaf. The results showed that TomatoGuard could be adopted as a model for identifying tomato diseases with a 99% test accuracy, showing significantly better performance compared with APP Plantix, a widely used APP for general purpose plant disease detection.

To optimize crop water consumption and adopt water-saving measures such as precision irrigation, early identification of plant water status is critical. This study explores the effectiveness of estimating water stress in choy sum (Brassica chinensis var. parachinensis) grown in pots in greenhouse conditions using Crop Water Stress Index (CWSI) and crop vegetation indicators to improve irrigation water management. Data on CWSI and Spectral reflectance were collected from choy sum plants growing in sandy loam soil with four different soil field capacities (FC): 90–100% FC as no water stress (NWS); 80–90% FC for light water stress (LWS); 70–80% FC for moderate water stress (MWS); and 60–70% FC for severe water stress (SWS). With four treatments and three replications, the experiment was set up as a completely randomized design (CRD). Throughout the growing season, plant water stress tracers such as leaf area index (LAI), canopy temperature (Tc), leaf relative water content (LRWC), leaf chlorophyll content, and yield were measured. Furthermore, CWSI estimated from the Workswell Wiris Agro R Infrared Camera (CWSIW) and spectral data acquisition from the Analytical Spectral Device on choy sum plants were studied at each growth stage. NDVI, Photochemical Reflectance Index positioned at 570 nm (PRI570), normalized PRI (PRInorm), Water Index (WI), and NDWI were the Vegetation indices (VIs) used in this study. At each growth stage, the connections between these CWSIW, VIs, and water stress indicators were statistically analyzed with R2 greater than 0.5. The results revealed that all VIs were valuable guides for diagnosing water stress in choy sum. CWSIW obtained from this study showed that Workswell Wiris Agro R Infrared Camera mounted on proximal remote sensing platform for assessing water stress in choy sum plant was rapid, non-destructive, and user friendly. Therefore, integrating CWSIW and VIs approach gives a more rapid and accurate approach for detecting water stress in choy sum grown under greenhouse conditions to optimize yield by reducing water loss and enhancing food security and sustainability.

The use of plant-based indicators and other conventional means to detect the level of water stress in crops may be challenging, due to their difficulties in automation, their arduousness, and their time-consuming nature. Non-contact and non-destructive sensing methods can be used to detect the level of water stress in plants continuously and to provide automatic sensing and controls. This research aimed at determining the viability, efficiency, and swiftness in employing the commercial Workswell WIRIS Agro R infrared camera (WWARIC) in monitoring water stress and scheduling appropriate irrigation regimes in mandarin plants. The experiment used a four-by-three randomized complete block design with 80–100% FC water treatment as full field capacity and three deficit irrigation treatments at 70–75% FC, 60–65% FC, and 50–55% FC. Air temperature, canopy temperature, and vapor pressure deficits were measured and employed to deduce the empirical crop water stress index, using the Idso approach (CWSI(Idso)) as well as baseline equations to calculate non-water stress and water stressed conditions. The relative leaf water content (RLWC) of mandarin plants was also determined for the growing season. From the experiment, CWSI(Idso) and CWSI were estimated using the Workswell Wiris Agro R infrared camera (CWSIW) and showed a high correlation (R2 = 0.75 at p < 0.05) in assessing the extent of water stress in mandarin plants. The results also showed that at an altitude of 12 m above the mandarin canopy, the WWARIC was able to identify water stress using three modes (empirical, differential, and theoretical). The WWARIC’s color map feature, presented in real time, makes the camera a suitable device, as there is no need for complex computations or expert advice before determining the extent of the stress the crops are subjected to. The results prove that this novel use of the WWARIC demonstrated sufficient precision, swiftness, and intelligibility in the real-time detection of the mandarin water stress index and, accordingly, assisted in scheduling irrigation.

The efficient use of water and fertilizer is vital for optimizing plant growth and yield in rice production. To achieve sustainable rice production and resource management, the ways in which applied water and nitrogen affect the root and shoot morpho-physiology, as well as yield, must be understood. In this study, a pot experiment was conducted to investigate the effects of slow-release nitrogen fertilizer (sulfur-coated urea) application at three levels (light nitrogen (NL), medium nitrogen (NM), and heavy nitrogen (NH)) on the growth, yield, and nitrogen use efficiency (NUE) of rice grown under three water regimes (wetting and soil saturation (WSS), wetting and moderate drying (WMD) and wetting and severe drying (WSD)). The results revealed that differences in water regimes and fertilizer rates led to significant differences in the roots, shoots, yield, and NUE of rice. Increasing the N dosage by 5% enhanced the root and biomass production by 16% in comparison with that of the other groups. The NH×WSS treatment produced the greatest root length, weight, density, active absorption, and oxidation. However, the integration of WSS × NL generated the maximum value of nitrogen apparent recovery efficiency (63.1% to 67.6%) and the greatest value of nitrogen partial factor productivity (39.9 g g−1 to 41.13 g g−1). Transmission electron microscopy (TEM) images showed that plants grown under high and medium nitrogen fertilizer rates with WSS had improved leaf mesophyll structure with normal starch grains, clear cell walls, and well-developed chloroplasts with tidy and well-arranged thylakoids. These results show that TEM images are useful for characterizing the nitrogen and water status of leaves in the sub-micrometer range and providing specific information regarding the leaf microstructure. The findings of this study suggest that the application of NH×WSS can produce improvements in growth traits and increase rice yield; however, the NL×WSS treatment led to greater NUE, and the authors recommend its usage in rice agriculture.

The microbial population (MP) is considered to be a relatively important part of soil health, quality, and productivity. Therefore, this study aimed to access the effects of tillage and rice cultivation modes on soil MP in Stagnic Anthrosols and Lateritic Red Earth soils. The treatments were as follows: (i) MDS: land tilled twice with a moldboard plow and hill-seeding of pregerminated seeds with a direct seeding machine of four to six seeds per hill at a planting space of 25 × 15 cm, (ii) RDS: land tilled twice with a rotary tiller and hill-seeding of pregerminated seeds with a direct seeding machine of four to six seeds per hill at a planting space of 25 × 15 cm, (iii) MMT: land tilled twice with a moldboard plow and 15-day-old seedlings were mechanically transplanted with a transplanting machine at a transplanting hill of four to six seedlings and at a transplanting space of 25 × 15 cm, and (iv) RMT: land tilled twice with a rotary tiller and 15-day-old seedlings were mechanically transplanted with a transplanting machine at a transplanting hill of four to six seedlings and a transplanting space of 25 × 15 cm. The findings showed that MDS improved the MP and increases rice yield. MDS showed a high increase in MP in both locations and the rice productivity of 32.81% (1H; first harvest) and 13.91% (2H; second harvest) and 16.48% (1H) and 18.13% (2H) for Zeng-Cheng and Yi-Yang, respectively. In conclusion, MDS was found to be better in improving the MP and increasing rice yield and could be adopted as a suitable approach for improving soil health, quality, and productivity.

Puddling is an important operation to minimize soil nutrient leaching and thereby increasing the availability of plant nutrients and achieving reduced soil condition. Good puddle field conditions are needed to create a favorable environment for normal growth of rice plants. However, long-term effects of puddling could lead to forms of large clods in fine textured soils, resulting in negative effect on the soil characteristics, preventing seed-soil contacts, and leading to decline in rice yield. This study was conducted in 2 years with treatment including puddling the land twice with moldboard plow and pregerminated seeds were hill-seeded with direct seeding machine (PD), puddling the land twice with rotary tiller and pregerminated seeds were hill-seeded with direct seeding machine (RD), puddling the land twice with moldboard plow and 15-day-old seedlings were hill-transplanted with transplant machine (PT), and puddling the land twice with rotary tiller and 15-day-old seedlings were hill-transplanted with transplant machine (RT) to assess the effects of puddling and rice establishment method on soil characteristics and rice yield. Results revealed significant improvement in the bulk density and increase in SOC, N, P, and K in PD. The maximum microbial population was found in PD. Rice grain yield showed a higher productivity increase of 7.65 t·ha−2 (31.25%) and 3.93 t·ha−2 (13.91%) for 1H and 2H (1H: 1st harvest and 2H: 2nd harvest), respectively, in PD compared with the lowest of 5.76 t·ha−1 and 3.45 t·ha−1 in 1H and 2H, respectively, under RT. Overall, PD was found to be the most suitable puddling type and rice establishment method for soil improvement and increasing rice yield.

The most important biotic stress factor impacting tomato crop biophysical, biochemical, physiological, and morphological features is water stress. A pot experiment was undertaken in a greenhouse to study the drought responsiveness of tomato (Solanum lycopersicum) yield and quality indices in sandy loam and silty loam soils. For both sandy loam and silty loam soils, the water supply levels were 70–100% FC, 60–70% FC, 50–60% FC, and 40–50% FC of ETo (crop evapotranspiration) from the vegetative stage to the fruit ripening stage, calculated using the Hargreaves–Samani (HS) model compared to the time-domain reflectometer (TDR) values calibrated using volumetric water content (VWC). The experiment was conducted as a 2 × 4 factorial experiment, arranged in a completely randomized block design, with four treatments replicated four times. In this study, we examined how sandy loam and silty loam soils at different % FC affect the total marketable yield and quality components of tomatoes, concentrating on total soluble solids (Brix), fruit firmness, dry fruit mass, pH, titratable acid (TA), ascorbic acid (Vit. C), and carotenoid composition. Lycopene and β-Carotene were estimated using the UV spectroscopy method, with absorption spectra bands centered at 451 nm, 472 nm, 485 nm, and 502 nm. The results revealed that even though there were some limitations, TDR-based soil moisture content values had a strong positive correlation with HS-based evapotranspiration, with R2 = 0.8, indicating an improvement whereby TDR can solely be used to estimate soil water content. Tomato plants subjected to 40–50% FC (ETo) water stress in both sandy loam and silty loam soils recorded the highest total soluble solids, titratable acidity, ascorbic acid content, and β-carotene content at an absorption peak of 482 nm, and lycopene content at an absorption peak of 472 nm, with lower fruit firmness, fruit juice content, and fruit juice pH, and a reduced marketable yield. Similarly, tomato plants subjected to 60–70% FC throughout the growing season achieved good fruit firmness, percent juice content, total soluble solids, titratable acidity, ascorbic acid content, and chlorophyll content (SPAD), with minimum fruit juice pH and high marketable yield in both soil textural types. It is concluded that subjecting tomato plants to 60–70% FC (ETo) has a constructive impact on the marketable yield quality indices of tomatoes.

Regulated deficit irrigation (RDI) is considered among the best water-saving techniques forsupplementing Regulated water to fully achieve the water needs of the plant while maximizing water productivity with little or no substantial decrease in final produce compared to the conventional forms of irrigating crops.The aim of this paper is to review existing RDI approaches used in citrus production as well as plant-water stress indicators.Most of the approaches employed in citrus RDI scheduling require weather data for evapotranspiration calculations which is very technical, laborious and time consuming.Nonetheless, the time domain reflectometer (TDR) offers a simple way of scheduling RDI based on the soil-water status at any given time.This approach will help address the challenges in setting up on-farm synoptic stations to measure weather data to compute evapotranspiration or from using data from weather stations which might be different from the farm conditions.The pros and cons of all the approaches have been discussed and recommended that the TDR can be adopted as an alternative to schedule irrigation in citrus orchards to ensure that plants are supplied with adequate volume of water for maximum water use efficiency.

Why puddling? It is an important operation to minimize soil nutrient leaching and thereby increasing the availability of plant nutrients and achieving reduced soil condition. Good puddle field conditions are needed to create favorable environment for normal growth of rice plants. However, long-term effects of puddling could lead to forms of large clods in fine textured soils; resulting in negative effect on the soil characteristics, preventing seed-soil contacts and leading to decline in rice yield. This study was conducted in 2 years with treatment including; puddling the land twice with moldboard plow and pre-germinated seeds were hill-seeded with direct seeding machine (PD), puddling the land twice with rotary tiller and pre-germinated seeds were hill-seeded with direct seeding machine (RD), puddling the land twice with moldboard plow and 15-day-old seedlings were hill-transplanted with transplant machine (PT), and puddling the land twice with rotary tiller and 15-day-old seedlings were hill-transplanted with transplant machine (RT) to assess the effect of puddling, direct seeding, and mechanical transplanting on soil characteristics and rice yield. Results revealed significant improvement in the bulk density and increase in SOC, N, P and K in PD. The maximum microbial population was found in PD. Rice yield showed a higher productivity increase of 7.44 t∙ha-2 and 3.91 t∙ha-2 in 2017, and 7.85 t∙ha-2 and 3.94 t∙ha-2 in 2018 respectively for 1H and 2H (1H: 1st harvest, and 2H: 2nd harvest) in PD. Overall, PD was found to be the most suitable puddling and rice establishment approach under paddy fields for soil improvement and increasing rice yield.

The goal of this research is to use a WORKSWELL WIRIS AGRO R INFRARED CAMERA (WWARIC) to assess the crop water stress index (CWSIW) on tomato growth in two soil types. This normalized index (CWSI) can map water stress to prevent drought, mapping yield, and irrigation scheduling. The canopy temperature, air temperature, and vapor pressure deficit were measured and used to calculate the empirical value of the CWSI based on the Idso approach (CWSIIdso). The vegetation water content (VWC) was also measured at each growth stage of tomato growth. The research was conducted as a 2 × 4 factorial experiment arranged in a Completely Randomized Block Design. The treatments imposed were two soil types: sandy loam and silt loam, with four water stress treatment levels at 70–100% FC, 60–70% FC, 50–60% FC, and 40–50% FC on the growth of tomatoes to assess the water stress. The results revealed that CWSIIdso and CWSIW proved a strong correlation in estimating the crop water status at R2 above 0.60 at each growth stage in both soil types. The fruit expansion stage showed the highest correlation at R2 = 0.8363 in sandy loam and R2 = 0.7611 in silt loam. VWC and CWSIW showed a negative relationship with a strong correlation at all the growth stages with R2 values above 0.8 at p < 0.05 in both soil types. Similarly, the CWSIW and yield also showed a negative relationship and a strong correlation with R2 values above 0.95, which indicated that increasing the CWSIW had a negative effect on the yield. However, the total marketable yield ranged from 2.02 to 6.8 kg plant−1 in sandy loam soil and 1.75 to 5.4 kg plant−1 in silty loam soil from a low to high CWSIW. The highest mean marketable yield was obtained in sandy loam soil at 70–100% FC (0.0 < CWSIW ≤ 0.25), while the least-marketable yield was obtained in silty loam soil 40–50% FC (0.75 < CWSIW ≤ 1.0); hence, it is ideal for maintaining the crop water status between 0.0 < CWSIW ≤ 0.25 for the optimum yield. These experimental results proved that the WWARIC effectively assesses the crop water stress index (CWSIW) in tomatoes for mapping the yield and irrigation scheduling.

As an important form of urban water resource, urban artificial lakes are severely affected by rapid urbanization and interference from human activities. These small lakes are characterized by their unique irregular shape, fragile ecosystem, and relatively closed, stagnant waterbodies. However, few studies have focused on their hydrodynamics and water quality, in particular the restoration methods and mechanisms remaining unclear. The present study applied the MIKE 21 FM model to investigate the effects of water diversion on water quality in a typical urban artificial lake. By considering different flow arrangements, several model scenarios were set up to predict the impacts of water diversion on selected water quality parameter. The results showed that the effectiveness of water diversion was directly related to flow velocity, the relative position to the fresh water inlet, the amount and quality of fresh water and water remaining to be diluted, and the circulation direction of flow field. The inflow–outflow arrangement was the primary factor determining the flow field and NH3–N variation trends across the lake, and an increased discharge exhibited unequal effects in individual zones. Wind was also important for the formation of flow circulation and pollutant variation. Methods were proposed for enhancing water quality in urban small-scale lakes, including changing the way diversion projects are managed, improving the quality of diverted flow, enhancing flow fluidity, or utilizing wind effects and local topography.

Drought and water scarcity due to global warming, climate change, and social development have been the most death-defying threat to global agriculture production for the optimization of water and food security. Reflectance indices obtained by an Analytical Spectral Device (ASD) Spec 4 hyperspectral spectrometer from tomato growth in two soil texture types exposed to four water stress levels (70–100% FC, 60–70% FC, 50–60% FC, and 40–50% FC) was deployed to schedule irrigation and management of crops’ water stress. The treatments were replicated four times in a randomized complete block design (RCBD) in a 2 × 4 factorial experiment. Water stress treatments were monitored with Time Domain Reflectometer (TDR) every 12 h before and after irrigation to maintain soil water content at the desired (FC%). Soil electrical conductivity (Ec) was measured daily throughout the growth cycle of tomatoes in both soil types. Ec was revealing a strong correlation with water stress at R2 above 0.95 p < 0.001. Yield was measured at the end of the end of the growing season. The results revealed that yield had a high correlation with water stress at R2 = 0.9758 and 0.9816 p < 0.01 for sandy loam and silty loam soils, respectively. Leaf temperature (LT °C), relative leaf water content (RLWC), leaf chlorophyll content (LCC), Leaf area index (LAI), were measured at each growth stage at the same time spectral reflectance data were measured throughout the growth period. Spectral reflectance indices used were grouped into three: (1) greenness vegetative indices; (2) water overtone vegetation indices; (3) Photochemical Reflectance Index centered at 570 nm (PRI570), and normalized PRI (PRInorm). These reflectance indices were strongly correlated with all four water stress indicators and yield. The results revealed that NDVI, RDVI, WI, NDWI, NDWI1640, PRI570, and PRInorm were the most sensitive indices for estimating crop water stress at each growth stage in both sandy loam and silty loam soils at R2 above 0.35. This study recounts the depth of 858 to 1640 nm band absorption to water stress estimation, comparing it to other band depths to give an insight into the usefulness of ground-based hyperspectral reflectance indices for assessing crop water stress at different growth stages in different soil types.

随着精准农业的发展,对农作物病害进行快速准确地识别是提高农作物产量、推动农业现代化的重要手段.传统的作物病害识别技术存在一定局限性,依赖人工提取特征,图像分割难度较大,特别在复杂环境下的识别效果不佳.而随着深度学习中卷积神经网络(Convolutional Neural Network,下文简称CNN)在图像识别领域上的不俗表现,不少研究人员将CNN应用于农作物的病害识别.研究表明,基于CNN的深度学习方法是目前对农作物病害分类识别最先进、最有效的方法.本文将阐述传统病害识别技术的原理和缺陷,并详细介绍了CNN技术原理及其在农作物病害识别中的应用案例,基于CNN的农作物病害识别在未来发展和应用上提出几点展望.

生态挡墙在河道整治中应用越来越广泛,但后期的植被管护因经费等原因往往达不到设计要求,为了减少人工管护,该文设计了由单片机、湿度传感器、太阳能板、铅酸蓄电池和水泵等的自动灌溉系统.该系统以太阳能为动力,由单片机通过温度传感器采集土壤湿度,再根据设定的土壤水分参数控制水泵工作进行微喷灌,以达到智能化灌溉,为生态挡墙植被管护提供了可行方案.

随着现代农业的进步,智能农业得到快速发展.传统农业灌溉大多采用大水漫灌的方式,水资源浪费严重,为了节约水资源和减少人力投入,智能灌溉系统成为研究热点.由于各地的农业生产条件不同,对灌溉系统的需求也不同.本文针对广东省人多地少和丘陵山区用电困难的现状,以单片机为核心,设计了基于太阳能的自动灌溉系统.该系统由单片机系统板、人机界面、土壤湿度传感器、太阳能板、铅酸蓄电池、充电控制器和直流水泵组成.湿度传感器将收集到的信息传给单片机,再由单片机控制水泵开关进行灌溉,实现智能化灌溉的需求,为山区农业发展提供参考.

为实现在自然环境条件下对苦瓜叶部病害的目标检测,该研究提出了一种基于改进的更快速区域卷积神经网络(Faster Region with Convolutional Neural Network Features,Faster R-CNN)的苦瓜叶部病害目标检测方法.Faster R-CNN以残差结构卷积神经网络ResNet-50作为该次试验的特征提取网络,将其所得特征图输入到区域建议网络提取区域建议框,并且结合苦瓜叶部病害尺寸小的特点,对原始的Faster R-CNN进行修改,增加区域建议框的尺寸个数,并在ResNet-50的基础下融入了特征金字塔网络(Feature Pyramid Networks,FPN).结果表明,该方法训练所得的深度学习网络模型具有良好的鲁棒性,平均精度均值(Mean Average Precision,MAP)值为78.85％;融入特征金字塔网络后,所得模型的平均精度均值为86.39％,提高了7.54％,苦瓜健康叶片、白粉病、灰斑病、蔓枯病、斑点病的平均精确率(Average Precision,AP)分别为89.24％、81.48％、83.31％、88.62％和89.28％,在灰斑病检测精度上比之前可提高了16.56％,每幅图像的检测时间达0.322 s,保证检测的实时性.该方法对复杂的自然环境下的苦瓜叶部病害检测具有较好的鲁棒性和较高的精度,对瓜果类疾病预防有重要的研究意义.