Search Results (48)

To address the challenges of low recognition accuracy and substantial model complexity in crop disease identification models operating in complex field environments, this study proposed a novel hybrid model named ConvTransNet-S, which integrates Convolutional Neural Networks (CNNs) and transformers for crop disease identification tasks. Unlike existing hybrid approaches, ConvTransNet-S uniquely introduces three key innovations: First, a Local Perception Unit (LPU) and Lightweight Multi-Head Self-Attention (LMHSA) modules were introduced to synergistically enhance the extraction of fine-grained plant disease details and model global dependency relationships, respectively. Second, an Inverted Residual Feed-Forward Network (IRFFN) was employed to optimize the feature propagation path, thereby enhancing the model’s robustness against interferences such as lighting variations and leaf occlusions. This novel combination of a LPU, LMHSA, and an IRFFN achieves a dynamic equilibrium between local texture perception and global context modeling—effectively resolving the trade-offs inherent in standalone CNNs or transformers. Finally, through a phased architecture design, efficient fusion of multi-scale disease features is achieved, which enhances feature discriminability while reducing model complexity. The experimental results indicated that ConvTransNet-S achieved a recognition accuracy of 98.85% on the PlantVillage public dataset. This model operates with only 25.14 million parameters, a computational load of 3.762 GFLOPs, and an inference time of 7.56 ms. Testing on a self-built in-field complex scene dataset comprising 10,441 images revealed that ConvTransNet-S achieved an accuracy of 88.53%, which represents improvements of 14.22%, 2.75%, and 0.34% over EfficientNetV2, Vision Transformer, and Swin Transformer, respectively. Furthermore, the ConvTransNet-S model achieved up to 14.22% higher disease recognition accuracy under complex background conditions while reducing the parameter count by 46.8%. This confirms that its unique multi-scale feature mechanism can effectively distinguish disease from background features, providing a novel technical approach for disease diagnosis in complex agricultural scenarios and demonstrating significant application value for intelligent agricultural management. Full article

Pigs are susceptible to the deadly infectious disease known as African swine fever (ASF), which is brought on by the African swine fever virus (ASFV). As such, prompt and precise disease detection is essential. Deletion of the virulence-related genes MGF-505/360 and EP402R generated from the virulent genotype II virus significantly reduces its virulence, and animal tests using one of the recombinant viruses show great lethality and transmissibility in pigs. The isothermal technique known as recombinase polymerase amplification (RPA) is perfect for rapid in-field detection. To accurately identify ASFV MGF-505R gene-deleted mutants and assess the complex infection situation of ASF, RPA assays in conjunction with real-time fluorescent detection (real-time RPA assay) and lateral flow dipstick (RPA-LFD assay) were created. These innovative methods allow for the direct detection of ASFV from pigs, offering in-field pathogen detection, timely disease management, and satisfying animal quarantine requirements. The specific primers and probes were designed against conserved regions of ASFV B646L and MGF-505R genes. Using recombinant plasmid DNA containing ASFV MGF-505R gene-deleted mutants as a template, the sensitivity of both ASF real-time RPA and ASF RPA-LFD assays were demonstrated to be 10 copies per reaction within 20 min at 37 °C. Neither assay had cross-reactions with CSFV, PRRSV, PPV, PRV, ot PCV2, common viruses seen in pigs, indicating that these methods were highly specific for ASFV. The evaluation of the performance of ASFV real-time RPA and ASFV RPA-LFD assays with clinical samples (n = 453) demonstrated their ability to specifically detect ASFV or MGF-505R gene-deleted mutants in samples of pig feces, ham, fresh pork, and blood. Both assays exhibited the same diagnostic rate as the WOAH-recommended real-time fluorescence PCR, highlighting their reliability and validity. These assays offer a simple, cost-effective, rapid, and sensitive method for on-site identification of ASFV MGF-505R gene-deleted mutants. As a promising alternative to real-time PCR, they have the potential to significantly enhance the prevention and control of ASF in field settings. Full article

As of today, bacteriological identification and the molecular approach PCR are considered the gold standards for Salmonella spp. detection. However, these methods are time-consuming and costly due to the requirements for enrichment and nucleic acid extraction. In this study, we evaluated the reliability of a developed colorimetric loop-mediated isothermal amplification (cLAMP) assay targeting the hilA gene, using Phenol Red as an amplification indicator. Given that Phenol Red is pH-dependent, and to develop an extraction-free test, we evaluated chicken meat pretreatment and thermal treatment. First, we assessed the reliability of this test using a pure culture of Salmonella spp. and then in 50 chicken samples pretreated with optimal NaOH concentrations under standardized conditions. Samples representing extreme pH values were artificially contaminated and subjected to DNA extraction and a heat-treatment protocol. Serial dilutions of these products served as templates for LAMP reactions. The assay sensitivity was estimated to be around 3.9 CFU/µL of pure bacterial culture. In contrast, in biological samples, we detected up to 10 CFU/µL using DNA extraction, while heat treatment successfully amplified the initial solution and even some dilutions up to 10³ CFU/µL. In conclusion, our cLAMP assay demonstrated good sensitivity and provided clear evidence of its potential for in-field use without relying on prior enrichment steps and DNA extraction. Full article

The systematic identification of appropriate sites for different rainwater harvesting (RWH) structures may contribute to better success of crop production in such areas. One approach to improving crop yields in North Kordofan, Sudan, that is mostly adaptable to the changing climate is in-field water harvesting. The main objective of this study is to employ a geographical information system (GIS) in order to identify the most suitable sites for setting in situ water harvesting structures, aiming to address climate change in this area. A GIS-based model was developed to generate suitability maps for in situ RWH using multi-criteria evaluation. Five suitability criteria (soil texture, runoff depth, rainfall surplus, land cover, and slope) were identified; then, five suitability levels were set for each criterion (excellent, good, moderate, poor, and unsuitable). Weights were assigned to the criteria based on their relative importance for RWH using the analytical hierarchy process (AHP). Using QGIS 2.6.1 and ArcGIS 10.2.2 software, all criterion maps and suitability maps were prepared. The obtained suitability map for the entire region showed that 40% of the region area fell within the “good” class, representing 7419.18 km², whereas 26% of the area was “excellent”, occupying 4863.75 km². However, only 8.9% and 15.6% of the entire region’s area were “poor” and “unsuitable” for RWH, respectively. The suitability map of the delineated pilot areas selected according to the attained FAO data revealed that one location, Wad_Albaga, was found to be in an excellent position, covering an area of 787.811 km², which represents 42.94% of the total area. In contrast, the Algabal location had 6.4% of its area classified as poor and the remaining portion classified as excellent. According to the findings from the validated trial, Wad_Albaga is located in a good site covering 844 km², representing 46.04%, while Algabal is classified as a moderate site, covering 341 km² or 18.6% of the area. This study concluded that the validation of the existing trial closely matched the suitability map derived using FAO data. However, ground data from field experiments provided more accurate results compared to the FAO suitability map. This study also concluded that using GIS is a time-saving and effective tool for identifying suitable sites and discovering the most appropriate locations for rainwater harvesting (RWH). Full article

The potential of loop-mediated isothermal amplification (LAMP) assays for species identification in medical diagnostics, food safety, and authentication is indisputable. The challenge in commercialization of such DNA-based rapid test methods for resource-limited settings is the on-site availability of the required reagents and an intuitive read-out system. In this work, reaction mixtures for LAMP assays for the detection of animal (plaice) and plant food (safflower) were lyophilized and stored at room temperature for up to 24 weeks, demonstrating that refrigeration of reagents after lyophilization is not absolutely necessary. During this period, reaction mixtures were stable and the polymerase showed no loss of activity. In addition, mobile testing, including DNA isolation, using the lyophilized LAMP reaction mixtures and a handheld fluorescence detection read-out system (Doctor Vida Pocket test) was successfully performed outside of the laboratory environment in less than 40 min using a proposed standard operation procedure. The results highlight that the use of the lyophilized LAMP reaction for food control purposes has the potential to meet the WHO-proposed REASSURED criteria. Full article

To optimize rice yield and enhance quality through targeted field management at each growth stage, rapid and accurate identification of rice growth stages is crucial. This study presents the Mobilenetv3-YOLOv8 rice growth-stage recognition model, designed for high efficiency and accuracy using Unmanned Aerial Vehicle (UAV) imagery. A UAV captured images of rice fields across five distinct growth stages from two altitudes (3 m and 20 m) across two independent field experiments. These images were processed to create training, validation, and test datasets for model development. Mobilenetv3 was introduced to replace the standard YOLOv8 backbone, providing robust small-scale feature extraction through multi-scale feature fusion. Additionally, the Coordinate Attention (CA) mechanism was integrated into YOLOv8’s backbone, outperforming the Convolutional Block Attention Module (CBAM) by enhancing position-sensitive information capture and focusing on crucial pixel areas. Compared to the original YOLOv8, the enhanced Mobilenetv3-YOLOv8 model improved rice growth-stage identification accuracy and reduced the computational load. With an input image size of 400 × 400 pixels and the CA implemented in the second and third backbone layers, the model achieved its best performance, reaching 84.00% mAP and 84.08% recall. The optimized model achieved parameters and Giga Floating Point Operations (GFLOPs) of 6.60M and 0.9, respectively, with precision values for tillering, jointing, booting, heading, and filling stages of 94.88%, 93.36%, 67.85%, 78.31%, and 85.46%, respectively. The experimental results revealed that the optimal Mobilenetv3-YOLOv8 shows excellent performance and has potential for deployment in edge computing devices and practical applications for in-field rice growth-stage recognition in the future. Full article

Tree- and block-level prediction of mango yield is important for farm operations, but current manual methods are inefficient. Previous research has identified the accuracies of mango yield forecasting using very-high-resolution (VHR) satellite imagery and an ’18-tree’ stratified sampling method. However, this approach still requires infield sampling to calibrate canopy reflectance and the derived block-level algorithms are unable to translate to other orchards due to the influences of abiotic and biotic conditions. To better appreciate these influences, individual tree yields and corresponding canopy reflectance properties were collected from 2015 to 2021 for 1958 individual mango trees from 55 orchard blocks across 14 farms located in three mango growing regions of Australia. A linear regression analysis of the block-level data revealed the non-existence of a universal relationship between the 24 vegetation indices (VIs) derived from VHR satellite data and fruit count per tree, an outcome likely due to the influence of location, season, management and cultivar. The tree-level fruit count predicted using a random forest (RF) model trained on all calibration data produced a percentage root mean squared error (PRMSE) of 26.5% and a mean absolute error (MAE) of 48 fruits/tree. The lowest PRMSEs produced from RF-based models developed from location, season and cultivar subsets at the individual tree level ranged from 19.3% to 32.6%. At the block level, the PRMSE for the combined model was 10.1% and the lowest values for the location, seasonal and cultivar subset models varied between 7.2% and 10.0% upon validation. Generally, the block-level predictions outperformed the individual tree-level models. Maps were produced to provide mango growers with a visual representation of yield variability across orchards. This enables better identification and management of the influence of abiotic and biotic constraints on production. Future research could investigate the causes of spatial yield variability in mango orchards. Full article

Background/Objectives: To evaluate whether additional confirmatory transperineal mapping biopsy (TPMB) in men with localized prostate cancer (PCa) alters the treatment plan and outcome of partial gland ablation (PGA) using high-intensity focused ultrasound (HIFU). Methods: We retrospectively reviewed data from 96 patients who underwent PGA using HIFU between January 2020 and June 2022. After multiparametric magnetic resonance imaging (mpMRI), all men underwent transrectal ultrasound (TRUS)-guided, cognitive-targeted biopsy and systematic biopsy. Men eligible for PGA using HIFU first underwent confirmatory TPMB. Any changes in the treatment plan after TPMB were analyzed. Follow-up TRUS-guided biopsy was performed 1 year post-operatively to evaluate oncological outcomes. Clinically significant PCa (csPCa) was defined as Gleason grade (GG) ≥ 2. Results: Among all subjects, the median age (IQR) was 65.0 (60.0–72.0) years and the prostate-specific antigen level was 5.20 (3.71–7.81) ng/mL. The results of both TRUS-guided biopsy and TPMB led to a change in the treatment plan (from unilateral to bilateral PGA) for 13 (13.5%) patients. The 1-year follow-up TRUS-guided biopsy identified PCa in 13 (13.5%) patients, and csPCa in 7 (7.3%) patients. The infield- and outfield-positive rates were 8.3% (8/96) and 3.1% (3/96), respectively, for any PCa, and 3.1% (3/96) and 2.1% (2/96), respectively, for csPCa. Conclusions: Confirmatory TPMB results in better disease identification and localization, thereby affecting the treatment plan and improving oncological outcomes. Therefore, confirmatory TPMB should be considered to establish an appropriate strategy for patients with localized PCa eligible for PGA using HIFU. Full article

Inertial Measurement Units are widely used in various applications and, hardware-wise, they primarily consist of a tri-axial accelerometer and a tri-axial gyroscope. For low-end commercial employments, the low cost of the device is crucial: this makes MEMS-based sensors a popular choice in this context. However, MEMS-based transducers are prone to significant, non-uniform and environmental-condition-dependent systematic errors, that require frequent re-calibration to be eliminated. To this end, identification methods that can be performed in-field by non-expert users, without the need for high-precision or costly equipment, are of particular interest. In this paper, we propose an in-field identification procedure based on the Total Least Squares method for both tri-axial accelerometers and gyroscopes. The proposed identification model is linear and requires no prior knowledge of the parameters to be identified. It enables accelerometer calibration without the need for specific reference surface orientation relative to Earth’s gravity and allows gyroscope calibration to be performed independently of accelerometer data, without requiring the sensor’s sensitive axes to be aligned with the rotation axes during calibration. Experiments conducted on NXP sensors FXOS8700CQ and FXAS21002 demonstrated that using parameters identified by our method reduced cross-validation standard deviations by about two orders of magnitude compared to those obtained using manufacturer-provided parameters. This result indicates that our method enables the effective calibration of IMU sensor parameters, relying only on simple 3D-printed equipment and significantly improving IMU performance at minimal cost. Full article

American plum line pattern virus (APLPV) is the most infrequently reported Ilarvirus infecting stone fruit trees and is of sufficient severity to be classified as an EPPO quarantine A1 pathogen. In late spring, yellow line pattern symptoms were observed on leaves in a few flowering cherries (Prunus serrulata Lindl.) grown in a public garden in Northwest Italy. RNA extracts from twenty flowering cherries were submitted to Ilarvirus multiplex and APLPV-specific RT-PCR assays already reported or developed in this study. One flowering cherry (T22) with mixed prunus necrotic ringspot virus (PNRSV) and prune dwarf virus (PDV) infection also showed infection with APLPV. Blastn analysis of PCR products of the full coat protein (CP) and movement protein (MP) genes obtained from flowering cherry T22 showed 98.23% and 98.34% nucleotide identity with reference APLPV isolate NC_003453.1 from the USA. Then, a LAMP-specific assay was designed to facilitate the fast and low-cost identification of this virus either in the laboratory or directly in the field. The developed assay allowed not only the confirmation of APLPV (PSer22IT isolate) infection in the T22 flowering cherry but also the identification of APLPV in an asymptomatic flowering cherry tree (TL1). The LAMP assay successfully worked with crude flowering cherry extracts, obtained after manually shaking a single plant extract in the ELISA extraction buffer for 3–5 min. The developed rapid, specific and economic LAMP assay was able to detect APLPV using crude plant extracts rather that RNA preparation in less than 20 min, making it suitable for in-field detection. Moreover, the LAMP assay proved to be more sensitive in APLPV detection in flowering cherry compared to the specific one-step RT-PCR assay. The new LAMP assay will permit the estimation of APLPV geographic spread in the territory, paying particular attention to surrounding gardens and propagated flowering cherries in ornamental nurseries. Full article

This study investigates the application of an eNose (electrochemical sensory array) device as a rapid and cost-effective screening tool to detect increasingly prevalent counterfeit electronic cigarettes, and those to which potentially hazardous excipients such as vitamin E acetate (VEA) have been added, without the need to generate and test the aerosol such products are intended to emit. A portable, in-field screening tool would also allow government officials to swiftly identify adulterated electronic cigarette e-liquids containing illicit flavorings such as menthol. Our approach involved developing canonical discriminant analysis (CDA) models to differentiate formulation components, including e-liquid bases and nicotine, which the eNose accurately identified. Additionally, models were created using e-liquid bases adulterated with menthol and VEA. The eNose and CDA model correctly identified menthol-containing e-liquids in all instances but were only able to identify VEA in 66.6% of cases. To demonstrate the applicability of this model to a commercial product, a Virginia Tobacco JUUL product was adulterated with menthol and VEA. A CDA model was constructed and, when tested against the prediction set, it was able to identify samples adulterated with menthol 91.6% of the time and those containing VEA in 75% of attempts. To test the ability of this approach to distinguish commercial e-liquid brands, a model using six commercial products was generated and tested against randomized samples on the same day as model creation. The CDA model had a cross-validation of 91.7%. When randomized samples were presented to the model on different days, cross-validation fell to 41.7%, suggesting that interday variability was problematic. However, a subsequently developed support vector machine (SVM) identification algorithm was deployed, increasing the cross-validation to 84.7%. A prediction set was challenged against this model, yielding an accuracy of 94.4%. Altered Elf Bar and Hyde IQ formulations were used to simulate counterfeit products, and in all cases, the brand identification model did not classify these samples as their reference product. This study demonstrates the eNose’s capability to distinguish between various odors emitted from e-liquids, highlighting its potential to identify counterfeit and adulterated products in the field without the need to generate and test the aerosol emitted from an electronic cigarette. Full article

Field zones at risk of low nitrogen use efficiency (NUE) can be identified by analyzing in-field spatial variability. This hypothesis was validated by analyzing soil mineral nitrogen (N_min) and several plant and soil N management indices. The research was conducted in Karmin (central Poland) during two growing seasons, with winter oilseed rape (2018/2019) and winter wheat (2019/2020). The study showed that the crop yield was positively related to N_min. However, this N trait did not explain all the observed differences in the spatial variation of crop yield and plant N accumulation. In addition, the soil N management indices were more spatially variable during the growing season than the plant N management indices. Particularly high variability was found for the indices characterizing the N surplus in the soil-plant system. The calculated N surplus (N_b = N fertilizer input − N seed output) ranged from −62.8 to 80.0 kg N ha⁻¹ (coefficient of variation, CV = 181.2%) in the rape field and from −123.5 to 8.2 kg N ha⁻¹ (CV = 60.2%) in the wheat field. The spatial distribution maps also confirm the high variability of the parameters characterizing the post-harvest N surplus, as well as the total N input (soil + fertilizer) to the field with rape. The results obtained indicate that a field N balance carried out in different field zones allows a more accurate identification of potential N losses from the soil-plant system. Full article

Crop identification at the field level using remote sensing data is a very important task. However, the use of multispectral data for the construction of vegetation indices is sometimes impossible or limited. For such situations, solutions based on the use of time series of synthetic aperture radar (SAR) indices are promising, eliminating the problems associated with cloudiness and providing an assessment of crop development characteristics during the growing season. We evaluated the use of time series of synthetic aperture radar (SAR) indices to characterize crop development during the growing season. The use of SAR imagery for crop identification addresses issues related to cloudiness. Therefore, it is important to choose the SAR index that is the most stable and has the lowest spatial variability throughout the growing season while being comparable to the normalized difference vegetation index (NDVI). The presented work is devoted to the study of these issues. In this study, the spatial variabilities of different SAR indices time series were compared for a single region for the first time to identify the most stable index for use in precision agriculture, including the in-field heterogeneity of crop sites, crop rotation control, mapping, and other tasks in various agricultural areas. Seventeen Sentinel-1B images of the southern part of the Khabarovsk Territory in the Russian Far East at a spatial resolution of 20 m and temporal resolution of 12 days for the period between 14 April 2021 and 1 November 2021 were obtained and processed to generate vertical–horizontal/vertical–vertical polarization (VH/VV), radar vegetation index (RVI), and dual polarimetric radar vegetation index (DpRVI) time series. NDVI time series were constructed from multispectral Sentinel-2 images using a cloud cover mask. The characteristics of time series maximums were calculated for different types of crops: soybean, oat, buckwheat, and timothy grass. The DpRVI index exhibited the highest stability, with coefficients of variation of the time series that were significantly lower than those for RVI and VH/VV. The main characteristics of the SAR and NDVI time series—the maximum values, the dates of the maximum values, and the variability of these indices—were compared. The variabilities of the maximum values and dates of maximum values for DpRVI were lower than for RVI and VH/VV, whereas the variabilities of the maximum values and the dates of maximum values were comparable for DpRVI and NDVI. On the basis of the DpRVI index, classifications were carried out using seven machine learning methods (fine tree, quadratic discriminant, Gaussian naïve Bayes, fine k nearest neighbors or KNN, random under-sampling boosting or RUSBoost, random forest, and support vector machine) for experimental sites covering a total area of 1009.8 ha. The quadratic discriminant method yielded the best results, with a pixel classification accuracy of approximately 82% and a kappa value of 0.67. Overall, 90% of soybean, 74.1% of oat, 68.9% of buckwheat, and 57.6% of timothy grass pixels were correctly classified. At the field level, 94% of the fields included in the test dataset were correctly classified. The paper results show that the DpRVI can be used in cases where the NDVI is limited, allowing for the monitoring of phenological development and crop mapping. The research results can be used in the south of Khabarovsk Territory and in neighboring territories. Full article

During the growing season, olives progress through nine different phenological stages, starting with bud development and ending with senescence. During their lifespan, olives undergo changes in their external color and chemical properties. To tackle these properties, we used hyperspectral imaging during the growing season of the olives. The objective of this study was to develop a lightweight model capable of identifying olives in the hyperspectral images using their spectral information. To achieve this goal, we utilized the hyperspectral imaging of olives while they were still on the tree and conducted this process throughout the entire growing season directly in the field without artificial light sources. The images were taken on-site every week from 9:00 to 11:00 a.m. UTC to avoid light saturation and glitters. The data were analyzed using training and testing classifiers, including Decision Tree, Logistic Regression, Random Forest, and Support Vector Machine on labeled datasets. The Logistic Regression model showed the best balance between classification success rate, size, and inference time, achieving a 98% F1-score with less than 1 KB in parameters. A reduction in size was achieved by analyzing the wavelengths that were critical in the decision making, reducing the dimensionality of the hypercube. So, with this novel model, olives in a hyperspectral image can be identified during the season, providing data to enhance a farmer’s decision-making process through further automatic applications. Full article

The wellbeing of the honey bee colonies and the health of humans are connected in numerous ways. Therefore, ensuring the wellbeing of bees is a crucial component of fostering sustainability and ecological harmony. The colony collapse disorder (CCD) phenomenon was first reported in 2006 when the majority of bee colonies in Europe died out, due to an increase in infections, contamination of hives with agrochemical pesticides, and persistent organic pollutants (POPs). Only 6 years after the emergence of CCD, more than 6.5 million premature deaths were reported, as a consequence of persistent human exposure to air pollution. The insect species such as the honey bee Apis mellifera L. and the air matrix inside the beehive can be used as tools in biomonitoring, instead of traditional monitoring methods. This may have advantages in terms of cost-effective bioindicators of the environmental health status, showing the ability to record spatial and temporal pollutant variations. In this study, we present the sustainable in-field usage of the portable membrane inlet mass spectrometry (MIMS) instrument for an instant and effective determination of the level of environmental pollution by analytical identification of hive atmosphere volatile organic compound (VOC) contaminants, polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs), monocyclic aromatic hydrocarbons (BTEX) compounds, and pesticides. The samples were taken from hives located in urbanized and rural regions, highlighting variations in contamination. The MIMS results were benchmarked against a conventional laboratory sampling technique, such as GC-MS. Full article