Search Results (61)

Frequent occurrences of pests and diseases in tomatoes severely restrict yield and quality improvements. Traditional detection methods are labor-intensive and prone to errors, while advancements in deep learning provide a promising solution for rapid and accurate identification. However, existing deep learning-based models often face high computational complexity and a large number of parameters, which hinder their deployment on resource-constrained edge devices. To overcome this limitation, we propose a novel lightweight detection model named YOLOv11n-KL based on the YOLOv11n framework. In this model, the feature extraction capability for small targets and the overall computational efficiency are enhanced through the integration of the Conv_KW and C3k2_KW modules, both of which incorporate the KernelWarehouse (KW) algorithm, and the Detect_LSCD detection head is employed to enable parameter sharing and adaptive multi-scale feature calibration. The results indicate that YOLOv11n-KL achieves superior performance in tomato pest and disease detection, balancing lightweight design and detection accuracy. The model achieves an mAP@0.5 of 92.5% with only 3.0 GFLOPs and 5.2 M parameters, reducing computational cost by 52.4% and improving mAP@0.5 by 0.9% over YOLOv11n. With its low complexity and high precision, YOLOv11n-KL is well-suited for resource-constrained applications. The proposed YOLOv11n-KL model offers an effective solution for detecting tomato pests and diseases, serving as a useful reference for agricultural automation. Full article

The olive psyllid (Euphyllura olivina), previously considered a secondary pest in Spain, is becoming more prevalent due to climate change and rising average temperatures. Its cottony wax secretions can cause substantial damage to olive crops under certain climatic conditions. Traditional monitoring methods for this pest are often labor-intensive, subjective, and impractical for large-scale surveillance. This study presents an automatic image segmentation approach based on deep learning to detect and quantify the cottony masses produced by E. olivina in olive trees. A well-annotated image dataset is developed and published, and a thorough evaluation of current camouflaged object detection (COD) methods is carried out for this task. Our results show that deep learning-based segmentation enables accurate and non-invasive assessment of pest symptoms, even in challenging visual conditions. However, further calibration and field validation are required before these methods can be deployed for operational integrated pest management. This work establishes a public dataset and a baseline benchmark, providing a foundation for future research and decision-support tools in precision agriculture. Full article

Unmanned aerial vehicles (UAVs) with multispectral sensors are transforming precision viticulture by enabling detailed monitoring of vineyard variability. Vegetation indices such as NDVI, NDRE, GNDVI, and SAVI are widely applied to estimate vine vigor, canopy structure, and water status. Beyond agronomic traits, UAV-derived indices can inform grape composition, including sugar content (°Brix), total phenolics, anthocyanins, titratable acidity, berry weight, and yield variables measurable in the field or laboratory to validate spectral predictions. Strengths of UAV approaches include high spatial resolution, rapid data acquisition, and flexibility across vineyard blocks, while limitations involve index saturation in dense canopies (e.g., Merlot, Cabernet Sauvignon), environmental sensitivity, and calibration requirements across varieties and phenological cycles. Integrating UAV data with ground-based measurements (leaf sampling, yield mapping, proximal or thermal sensors) improves model accuracy and stress detection. Abiotic stresses (water deficit, nutrient deficiency) can be distinguished from biotic factors (pest and fungal infections), supporting timely interventions. Compared to manned aircraft or satellite platforms, UAVs offer cost-effective, high-resolution imagery for precision vineyard management. Future directions include combining UAV indices with machine learning and data fusion to predict grape maturity and wine quality, enhancing decision-making in sustainable viticulture and precision enology. Full article

Hemp (Cannabis sativa L.) cultivation in the United States has expanded rapidly over the past decade. Due to federal and state regulations, only a limited number of studies have examined the chemical options available for controlling pests on C. sativa. In the U.S., two of the most important species of arthropod pests affecting C. sativa are the beet leafhopper Circulifer tenellus Baker (Hemiptera: Cicadellidae) and the two-spotted spider mite Tetranychus urticae Koch (Acari: Tetranychidae). This study evaluated the effects of four biopesticides, Chromobacterium subtsugae, Burkholderia spp., Chenopodium ambrosioides, and azadirachtin, under greenhouse conditions against C. tenellus adults and nymphs and T. urticae adults. Biopesticides were applied to foliage using a calibrated hand sprayer. To evaluate the biopesticides’ potency, C. tenellus adults, nymphs, and mites were released 1 h after treatment; to evaluate the residual efficacy, they were released 7 days after treatment (DAT). In both experiments, C. tenellus adults, nymphs, and mites were counted 1, 3, and 7 days after release. Our results indicate that Burkholderia spp. exhibited the highest efficacy against C. tenellus adults at 7 DAT, whereas C. ambrosioides and azadirachtin caused the greatest nymphal mortality at 1 and 3 DAT, respectively. Our results show that Burkholderia spp. had the greatest potency against C. tenellus adults 7 DAT, while C. ambrosioides and azadirachtin highly affect the mortality of nymphs at 1 and 3 DAT, respectively. Treatments with C. subtsugae and C. ambrosioides showed high potency against T. urticae. Finally, C. subtsugae showed the lowest residual effect against the mite pest. The data presented in this article will add to the arsenal of information to improve the current management strategies used against these two hemp pests. Full article

Accurate monitoring of orchard pests is vital for green and efficient apple production. Yet images captured by intelligent pest-monitoring lamps often contain small targets, weak boundaries, and crowded scenes, which hamper detection accuracy. We present MDAS-YOLO, a lightweight detection framework tailored for smart pest monitoring in apple orchards. At the input stage, we adopt the LIME++ enhancement to mitigate low illumination and non-uniform lighting, improving image quality at the source. On the model side, we integrate three structural innovations: (1) a C3k2-MESA-DSM module in the backbone to explicitly strengthen contours and fine textures via multi-scale edge enhancement and dual-domain feature selection; (2) an AP-BiFPN in the neck to achieve adaptive cross-scale fusion through learnable weighting and differentiated pooling; and (3) a SimAM block before the detection head to perform zero-parameter, pixel-level saliency re-calibration, suppressing background redundancy without extra computation. On a self-built apple-orchard pest dataset, MDAS-YOLO attains 95.68% mAP, outperforming YOLOv11n by 6.97 percentage points while maintaining a superior trade-off among accuracy, model size, and inference speed. Overall, the proposed synergistic pipeline—input enhancement, early edge fidelity, mid-level adaptive fusion, and end-stage lightweight re-calibration—effectively addresses small-scale, weak-boundary, and densely distributed pests, providing a promising and regionally validated approach for intelligent pest monitoring and sustainable orchard management, and offering methodological insights for future multi-regional pest monitoring research. Full article

The groundwater inverse problem has several challenges such as instability, non-uniqueness, and complexity, especially for heterogeneous aquifers. Solving the inverse problem is the traditional way to calibrate models, but it is both time-consuming and sensitive to errors in the measurements. This study explores the use of machine learning (ML) surrogate models, namely Random Forest (RF) and Extreme Gradient Boosting (XGBoost), to solve the inverse problem for the groundwater model calibration. Datasets for 20 hydraulic conductivity fields were created randomly based on statistics of hydraulic conductivity from the available data of the Northern Aquifer of Qatar, which was used as a case study. The corresponding hydraulic head values were obtained using MODFLOW simulations, and the data were used to train and validate the ML models. The trained surrogate models were used to estimate the hydraulic conductivity based on field observations. The results show that both RF and XGBoost have considerable predictive skill, with RF having better R² and RMSE values (R² = 0.99 for training, 0.93 for testing) than XGBoost (R² = 0.86 for training, 0.85 for testing). The ML-based method lowered the computational effort greatly compared to the classical solution of the inverse problem (i.e., using PEST) and still produced strong and reliable spatial patterns of hydraulic conductivity. This demonstrates the potential of machine learning models for calibrating complex groundwater systems. Full article

The worldwide agriculture industry is facing increasing problems due to rapid population increase and increasingly unfavorable weather patterns. In order to reach the projected food production targets, which are essential for guaranteeing global food security, innovative and sustainable agricultural methods must be adopted. Conventional approaches, including traditional breeding procedures, often cannot handle the complex and simultaneous effects of biotic pressures such as pest infestations, disease attacks, and nutritional imbalances, as well as abiotic stresses including heat, salt, drought, and heavy metal toxicity. Applying phytohormonal approaches, particularly those involving hormonal crosstalk, presents a viable way to increase crop resilience in this context. Abscisic acid (ABA), gibberellins (GAs), auxin, cytokinins, salicylic acid (SA), jasmonic acid (JA), ethylene, and GA are among the plant hormones that control plant stress responses. In order to precisely respond to a range of environmental stimuli, these hormones allow plants to control gene expression, signal transduction, and physiological adaptation through intricate networks of antagonistic and constructive interactions. This review focuses on how the principal hormonal signaling pathways (in particular, ABA-ET, ABA-JA, JA-SA, and ABA-auxin) intricately interact and how they affect the plant stress response. For example, ABA-driven drought tolerance controls immunological responses and stomatal behavior through antagonistic interactions with ET and SA, while using SnRK2 kinases to activate genes that react to stress. Similarly, the transcription factor MYC2 is an essential node in ABA–JA crosstalk and mediates the integration of defense and drought signals. Plants’ complex hormonal crosstalk networks are an example of a precisely calibrated regulatory system that strikes a balance between growth and abiotic stress adaptation. ABA, JA, SA, ethylene, auxin, cytokinin, GA, and BR are examples of central nodes that interact dynamically and context-specifically to modify signal transduction, rewire gene expression, and change physiological outcomes. To engineer stress-resilient crops in the face of shifting environmental challenges, a systems-level view of these pathways is provided by a combination of enrichment analyses and STRING-based interaction mapping. These hormonal interactions are directly related to the United Nations Sustainable Development Goals (SDGs), particularly SDGs 2 (Zero Hunger), 12 (Responsible Consumption and Production), and 13 (Climate Action). This review emphasizes the potential of biotechnologies to use hormone signaling to improve agricultural performance and sustainability by uncovering the molecular foundations of hormonal crosstalk. Increasing our understanding of these pathways presents a strategic opportunity to increase crop resilience, reduce environmental degradation, and secure food systems in the face of increasing climate unpredictability. Full article

A VOC-contaminated shallow aquifer in an industrial site was investigated to evaluate its potential for natural attenuation. The shallow groundwater aquifer beneath the industrial site has been contaminated by dissolved volatile organic compounds (VOCs) such as trichloroethylene (TCE), cis-1,2-dichloroethylene (cis-DCE), and vinyl chloride (VC) for more than three decades. Monitoring and investigation were implemented during 2011–2024, aiming to propose future groundwater aquifer management strategies. This study included groundwater borehole investigation, well installation monitoring, hydraulic head measurements, slug tests, groundwater samplings, and microbial analyses. Microbial investigations identified the predominant group of microorganisms of Proteobacteria, indicating biodegradation potential, as demonstrated by the presence of cis-DCE and VC. BIOSCREEN was used to evaluate the process of natural attenuation, incorporating site-specific parameters. A two-layer groundwater flow model was developed using MODFLOW with hydraulic conductivities obtained from slug tests. The site has an average hydraulic head of 259.6 m amsl with a hydraulic gradient of 0.026, resulting in an average groundwater flow velocity of 11 m/y. Hydraulic conductivities were estimated during model calibration using the PEST pilot point technique. A reactive transport model, RT3D, was used to simulate dissolved TCE transport over 30 years, which can undergo sorption as well as biodegradation. Model calibration demonstrated a satisfactory fit between observed and simulated groundwater heads with a root mean square error of 0.08 m and a correlation coefficient (r) between measured and simulated heads of 0.81, confirming the validity of the hydraulic conductivity distribution. The TCE plume continuously degraded and gradually migrated southward, generating a cis-DCE plume. The concentrations in both plumes decreased toward the end of the simulation period at Source 1 (located upstream), while BIOSCREEN results confirmed ongoing natural attenuation primarily by biodegradation. The integrated MODFLOW-RT3D-BIOSCREEN approach effectively evaluated VOC attenuation and plume migration. However, future remediation strategies should consider enhanced bioremediation to accelerate contaminant degradation at Source 2 and ensure long-term groundwater quality. Full article

Chinese cabbage (Brassica rapa) is one of the most important fall vegetables in South Korea. Recently, cabbage yields fluctuated due to climate change, leading to an unstable supply and increased prices. Additionally, raised temperatures led to increased beet armyworm (Spodoptera exigua) populations, resulting in greater plant damage. In this study, the Agricultural Policy/Environmental Extender (APEX) model was employed to develop the cabbage growth model. To enhance model accuracy, 4 years of field data collected from multiple locations in South Korea were utilized for model validation and calibration. The model goodness of fit tests revealed R² values between 0.9485 and 0.9873. Two different cabbage models, representing the physiological characteristics of common varieties cultivated in Korea, were applied to assess growth patterns under two distinct climate change scenarios, SSP245 and SSP585. A larval duration prediction model was formulated using previous field data. Under future climate conditions, simulation results indicate that as temperatures rise, Chinese cabbage yields will likely decrease continually, with increasing plant damage from insects. The modeling results can help farmers to control and manage crop insect pests under varying environmental conditions. Full article

The integration of machine learning (ML) into self-governing spraying systems is one of the major developments in digital precision agriculture that is significantly improving resource efficiency, sustainability, and production. This study looks at current advances in machine learning applications for automated spraying in agricultural mechanisation, emphasising the new innovations, difficulties, and prospects. This study provides an in-depth analysis of the three main categories of autonomous sprayers—drones, ground-based robots, and tractor-mounted systems—that incorporate machine learning techniques. A comprehensive review of research published between 2014 and 2024 was conducted using Web of Science and Scopus, selecting relevant studies on agricultural robotics, sensor integration, and ML-based spraying automation. The results indicate that supervised, unsupervised, and deep learning models increasingly contribute to improved real-time decision making, performance in pest and disease detection, as well as accurate application of agricultural plant protection. By utilising cutting-edge technology like multispectral sensors, LiDAR, and sophisticated neural networks, these systems significantly increase spraying operations’ efficiency while cutting waste and significantly minimising their negative effects on the environment. Notwithstanding significant advances, issues still exist, such as the requirement for high-quality datasets, system calibration, and flexibility in a range of field circumstances. This study highlights important gaps in the literature and suggests future areas of inquiry to develop ML-driven autonomous spraying even more, assisting in the shift to more intelligent and environmentally friendly farming methods. Full article

The production of viable nucleopolyhedrovirus in the feces of infected lepidopteran larvae represents a poorly understood route for virus transmission prior to host death. In the present study, we examined the presence of viable virus in the feces of fourth-instar Spodoptera frugiperda larvae infected with the Nicaraguan isolate of Spodoptera frugiperda multiple nucleopolyhedrovirus (SfMNPV-NIC). Feces production increased in samples taken at 2 to 6 days post-inoculation but was significantly lower in infected insects compared to controls. Second instars experienced 3.9 to 68.3% of polyhedrosis disease following consumption of feces collected at 2–5 days post-inoculation, which subsequently fell to 29.1% in the 6-day sample. Calibration of the insect bioassay using OB-spiked samples of feces indicated that the concentration of OBs varied between 5.4 × 10² and 4.4 × 10⁵ OBs/100 mg of feces in infected fourth instars. Quantitative PCR analysis of fecal samples indicated the presence of 0 to 7629 copies/mg feces following amplification targeted at the polyhedrin gene. However, no correlation was detected between qPCR estimates of virus concentration and time of sample collection or the quantity of feces collected. The qPCR estimates were positively correlated with the prevalence of lethal infection observed in the insect bioassay, but the correlation was weak and several samples did not amplify. Calibration of the qPCR assay using OB-spiked samples of feces provided estimates that were 5- to 10-fold lower than the insect bioassay, indicating inhibition of the amplification reaction or loss of material during processing. In a greenhouse experiment, 2.5–48.3% of second-instar larvae acquired lethal infection following a 24 h period of feeding on maize plants on which fourth instar larvae had deposited their feces at 3 days and 4 days post-infection, respectively. These findings highlight the potential of OB-contaminated feces as a source of biologically significant quantities of inoculum for virus transmission prior to the death of infected insects and represent an additional contribution to the biological control of lepidopteran pests by these pathogens. Full article

Grape phylloxera, Daktulosphaira vitifoliae (Fitch), is an economically significant pest of grapevines. Identification of phylloxera genotypes is an important aspect of management as genotypes differ in virulence and susceptibility to control using resistant rootstocks. Microsatellite markers developed on polyacrylamide gel systems have been the most widely used molecular method for phylloxera genotype identification, but this approach has been superseded by fluorescent capillary-based genotyping. The current study presents new laboratory methods for amplifying a standard set of eight phylloxera microsatellite markers using PCR-incorporated fluorescently labelled primers, genotyped on an ABI capillary platform. Comparison of allele size data scored on (i) polyacrylamide, (ii) capillary, and (iii) high-throughput sequencing (HTS) platforms revealed that the capillary genotyping most closely matched the HTS allele sizes, while alleles of loci originally scored on a polyacrylamide platform differ in size by up to three base pairs, mostly due to the presence of previously uncharacterised DNA sequence indels. Seven common clonal lineages of phylloxera known from Australia are proposed as reference samples for use in calibrating genotyping systems between platforms and laboratories to ensure universal scoring of allele sizes, providing a critical link for accurately matching previous phylloxera genotype studies with current research. Full article

The color of the plant leaves is a major concern in many areas of agriculture. Pigmentation and its pattern provide the possibility to distinguish genotypes and a basis for annual crop management practices. For example, the nutrient and water status of plants is reflected in the chlorophyll content of leaves that are strongly linked to the lamina coloration. Pests and diseases (virus or bacterial infections) also cause symptoms on the foliage. These symptoms induced by biotic and abiotic stressors often have a specific pattern, which allows for their prediction based on remote sensing. In this report, an RGB (red, green and blue) image processing system is presented to determine leaf lamina color variability based on RGB-based color indices. LeafLaminaMap was developed in Scilab with the Image Processing and Computer Vision toolbox, and the code is available freely at GitHub. The software uses RGB images to visualize 29 color indices and the R, G and B values on the lamina, as well as to calculate the statistical parameters. In this case study, symptomatic (senescence, fungal infection, etc.) and healthy grapevine (Vitis vinifera L.) leaves were collected, digitalized and analyzed with the LeafLaminaMap software according to the mean, standard deviation, contrast, energy and entropy of each channel (R, G and B) and color index. As an output for each original image in the sample set, the program generates 32 images, where each pixel is constructed using index values calculated from the RGB values of the corresponding pixel in the original image. These generated images can subsequently be used to help the end-user identify locally occurring symptoms that may not be visible in the original RGB image. The statistical evaluation of the samples showed significant differences in the color pattern between the healthy and symptomatic samples. According to the F value of the ANOVA analysis, energy and entropy had the largest difference between the healthy and symptomatic samples. Linear discriminant analysis (LDA) and support vector machine (SVM) analysis provided a perfect recognition in calibration and confirmed that energy and entropy have the strongest discriminative power between the healthy and symptomatic samples. The case study showed that the LeafLaminaMap software is an effective environment for the leaf lamina color pattern analysis; moreover, the results underline that energy and entropy are valuable features and could be more effective than the mean and standard deviation of the color properties. Full article

Stored products, such as grains and processed foods, are susceptible to infestation by various insects. The early detection of insects in the supply chain is crucial, as introducing invasive pests to new environments may cause disproportionate harm. The STAR Center at Stevens Institute of Technology developed the Acoustic Stored Product Insect Detection System (A-SPIDS) to detect pests in stored products. The system, which comprises a sound-insulated container for product samples with a built-in internal array of piezoelectric sensors and additional electret microphones to record outside noise, was used to conduct numerous measurements of the vibroacoustic signatures of various insects, including the Callosobruchus maculatus, Tribolium confusum, and Tenebrio molitor, in different materials. A normalization method was implemented using the ambient noise of the sensors as a reference, to accommodate for the proprietary, non-calibrated sensors and allowing to set relative detection thresholds for unknown sensitivities. The normalized envelope of the filtered signals was used to characterize and compare the insect signals by estimating the Normalized Signal Pulse Amplitude (NSPA) and the Normalized Spectral Energy Level (NSEL). These parameters characterize the insect detection Signal Noise Ratio (SNR) for pulse-based detection (NSPA) and averaged energy-based detection (NSEL). These metrics provided an initial step towards the design of a reliable detection algorithm. In the conducted tests NSPA was significantly larger than NSEL. The NSPA reached 70 $\mathrm{d}$$\mathrm{B}$ for T. molitor in corn flakes. The insect signals were lower in flour where the averaged NSPA and NSEL values were around 40 dB and 11 $\mathrm{d}$$\mathrm{B}$ to 16 $\mathrm{d}$$\mathrm{B}$, respectively. Full article

The Biome-BGC model is a classic forest ecosystem carbon cycle model driven by remote sensing and plant trait data, and it has been widely applied in various regions of China over the years. However, does the Biome-BGC model have good applicability in all regions of China? This question implies that the rationality of some applications of the Biome-BGC model in China might be questionable. To quantitatively assess the overall spatial applicability of the Biome-BGC model in China’s vegetation ecosystems, this study selected ten representative forest and grassland ecosystem sites, all of which have publicly available carbon flux data. In this study, we first used the EFAST method to identify the sensitive ecophysiological parameters of the Biome-BGC model at these sites. Subsequently, we calibrated the optimal values of these sensitive parameters through a literature review and the PEST method and then used these to drive the Biome-BGC model to simulate the productivity (including GPP and NEP) of these ten forest and grassland ecosystems in China. Finally, we compared the simulation accuracy of the Biome-BGC model at these ten sites in detail and established the spatial pattern of the model’s applicability across China. The results show that the sensitive ecophysiological parameters of the Biome-BGC model vary with spatial distribution, plant functional types, and model output variables. After conducting parameter sensitivity analysis and optimization, the simulation accuracy of the Biome-BGC model can be significantly improved. Additionally, for forest ecosystems in China, the model’s simulation accuracy decreases from north to south, while for grassland ecosystems, the accuracy increases from north to south. This study provides a set of localized ecophysiological parameters and advocates that the use of the Biome-BGC model should be based on parameter sensitivity analysis and optimization. Full article