Search Results (7,965)

Extraction of dense 3D geographic information from ultra-high-resolution unmanned aerial vehicle (UAV) imagery unlocks a great number of mapping and monitoring applications. This is facilitated by a step called dense image matching, which tries to find pixels corresponding to the same object within overlapping images captured by the UAV from different locations. Recent developments in deep learning utilize deep convolutional networks to perform this dense pixel correspondence task. A common theme in these developments is to train the network in a supervised setting using available dense 3D reference datasets. However, in this work we propose a novel unsupervised dense point cloud extraction routine for UAV imagery, called UnDER. We propose a novel disparity-shifting procedure to enable the use of a stereo matching network pretrained on an entirely different typology of image data in the disparity-estimation step of UnDER. Unlike previously proposed disparity-shifting techniques for forming cost volumes, the goal of our procedure was to address the domain shift between the images that the network was pretrained on and the UAV images, by using prior information from the UAV image acquisition. We also developed a procedure for occlusion masking based on disparity consistency checking that uses the disparity image space rather than the object space proposed in a standard 3D reconstruction routine for UAV data. Our benchmarking results demonstrated significant improvements in quantitative performance, reducing the mean cloud-to-cloud distance by approximately 1.8 times the ground sampling distance (GSD) compared to other methods. Full article

Identifying suitable sites for urban, industrial, and tourist development is important, especially in areas with increasing population and limited land availability. Kharga Oasis, Egypt, stands out as a promising area for such development, which can help reduce overcrowding in the Nile Valley and Delta. However, soil and various environmental factors can affect the suitability of civil engineering projects. This study used Geographic Information Systems (GISs) and a multi-criteria decision-making approach to assess the suitability of Kharga Oasis for construction activities. Geotechnical parameters were obtained from seismic velocity data, including Poisson’s ratio, stress ratio, concentration index, material index, N-value, and foundation-bearing capacity. A comprehensive analysis of in situ and laboratory-based geological and geotechnical data from 24 boreholes examined soil plasticity, water content, unconfined compressive strength, and consolidation parameters. By integrating geotechnical, geomorphological, geological, environmental, and field data, a detailed site suitability map was created using the analytic hierarchy process to develop a weighted GIS model that accounts for the numerous elements influencing civil project design and construction. The results highlight suitable sites within the study area, with high and very high suitability classes covering 56.87% of the land, moderate areas representing 27.61%, and unsuitable areas covering 15.53%. It should be noted that many settlements exist in highly vulnerable areas, emphasizing the importance of this study. This model identifies areas vulnerable to geotechnical and geoenvironmental hazards, allowing for early decision-making at the beginning of the planning process and reducing the waste of effort. The applied model does not only highlight suitable sites in the Kharga Oasis, Egypt, but, additionally, it provides a reproducible method for efficiently assessing land use suitability in other regions with similar geological and environmental conditions around the world. Full article

In today’s industrial landscape, optimizing energy consumption, reducing production times, and maintaining quality standards are critical challenges, particularly in energy-intensive processes like resistance spot welding (RSW). This study introduces an intelligent decision support system designed to optimize the RSW process for steel reinforcement bars. By creating robust machine learning models trained on limited datasets, the system generates interactive heat maps that provide real-time guidance to production engineers or intelligent systems, enabling dynamic adaptation to changing conditions and external factors such as fluctuating energy costs. These heat maps serve as a flexible and intuitive tool for identifying robust operational points that balance quality, energy efficiency, and productivity. The proposed methodology advances decision-making in welding processes by combining robust predictive modeling with innovative visualization techniques, offering a versatile solution for multiobjective optimization in real-world industrial applications. Full article

In plasticulture production systems, the conventional practice involves broadcasting pre-emergent herbicides over the entire surface of raised beds before laying plastic mulch. However, weed emergence predominantly occurs through the transplant punch-holes in the mulch, leaving most of the applied herbicide beneath the plastic, where weeds cannot grow. To address this issue, we developed and evaluated a precision spraying system designed to target herbicide application to the transplant punch-holes. A dataset of 3378 images was manually collected and annotated during a tomato experimental trial at the University of Florida. A YOLOv8x model with a p2 output layer was trained, converted to TensorRT^® to improve the inference time, and deployed on a custom-built computer. A Python-based graphical user interface (GUI) was developed to facilitate user interaction and the control of the smart sprayer system. The sprayer utilized a global shutter camera to capture real-time video input for the YOLOv8x model, which activates or disactivates a TeeJet solenoid for precise herbicide application upon detecting a punch-hole. The model demonstrated excellent performance, achieving precision, recall, mean average precision (mAP), and F1score exceeding 0.90. Field tests showed that the smart sprayer reduced herbicide use by up to 69% compared to conventional broadcast methods. The system achieved an 86% punch-hole recognition rate, with a 14% miss rate due to challenges such as plant occlusion and variable lighting conditions, indicating that the dataset needs to be improved. Despite these limitations, the smart sprayer effectively minimized off-target herbicide application without causing crop damage. This precision approach reduces chemical inputs and minimizes the potential environmental impact, representing a significant advancement in sustainable plasticulture weed management. Full article

Detecting binary changes in co-registered bitemporal hyperspectral images (HSIs) using deep learning methods is challenging due to the high dimensionality of spectral data and significant variations between images. To address this challenge, previous approaches often used dimensionality reduction methods separately from the change detection network, leading to less accurate results. In this study, we propose an end-to-end fully connected adversarial network (EFC-AdvNet) for binary change detection, which efficiently reduces the dimensionality of bitemporal HSIs and simultaneously detects changes between them. This is accomplished by extracting critical spectral features at the pixel level through a self-spectral reconstruction (SSR) module working in conjunction with an adversarial change detection (Adv-CD) module to effectively delineate changes between bitemporal HSIs. The SSR module employs a fully connected autoencoder for hyperspectral dimensionality reduction and spectral feature extraction. By integrating the encoder segment of the SSR module with the change detection network of the Adv-CD module, we create a generator that directly produces highly accurate change maps. This joint learning approach enhances both feature extraction and change detection capabilities. The proposed network is trained using a comprehensive loss function derived from the concurrent learning of the SSR and Adv-CD modules, establishing EFC-AdvNet as a robust end-to-end network for hyperspectral binary change detection. Experimental evaluations of EFC-AdvNet on three public hyperspectral datasets demonstrate that joint learning between the SSR and Adv-CD modules improves the overall accuracy (OA) by 5.44%, 10.43%, and 7.52% for the Farmland, Hermiston, and River datasets, respectively, compared with the separate learning approach. Full article

Accurate counting of the number of rice panicles per unit area is essential for rice yield estimation. However, intensive planting, complex growth environments, and the overlapping of rice panicles and leaves in paddy fields pose significant challenges for precise panicle detection. In this study, we propose YOLO-Rice, a rice panicle detection model based on the You Only Look Once version 8 nano (YOLOv8n). The model employs FasterNet, a lightweight backbone network, and incorporates a two-layer detection head to improve rice panicle detection performance while reducing the overall model size. Additionally, we integrate a Normalization-based Attention Module (NAM) and introduce a Minimum Point Distance-based IoU (MPDIoU) loss function to further improve the detection capability. The results demonstrate that the YOLO-Rice model achieved an object detection accuracy of 93.5% and a mean Average Precision (mAP) of 95.9%, with model parameters reduced to 32.6% of the original YOLOv8n model. When deployed on a Raspberry Pi 5, YOLO-Rice achieved 2.233 frames per second (FPS) on full-sized images, reducing the average detection time per image by 81.7% compared to YOLOv8n. By decreasing the input image size, the FPS increased to 11.36. Overall, the YOLO-Rice model demonstrates enhanced robustness and real-time detection capabilities, achieving higher accuracy and making it well-suited for deployment on low-cost portable devices. This model offers effective support for rice yield estimation, as well as for cultivation and breeding applications. Full article

The targets in the aerial view of UAVs are small, scenes are complex, and background noise is strong. Additionally, the low computational capability of UAVs is challenged when trying to meet the requirements of large neural networks. Therefore, a lightweight object detection algorithm tailored for UAV platforms, called RSG-YOLO, is proposed. The algorithm introduces an attention module constructed with receptive field attention and coordinate attention, which helps reduce background noise interference while improving long-range information dependency. It also introduces and refines a fine-grained downsampling structure to minimize the loss of target information during the downsampling process. A general efficient layer aggregation network enhances the base feature extraction module, improving gradient flow information. Additionally, a detection layer rich in small target information is added, while redundant large object detection layers are removed, achieving a lightweight design while enhancing detection accuracy. Experimental results show that, compared to the baseline algorithm, the improved algorithm increases the P, R, [email protected], and [email protected]:0.95 by 6.9%, 7.2%, 8.4%, 5.8%, respectively, on the VisDrone 2019 dataset, and by 5.7%, 9%, 9.3%, 3.6%, respectively, on the TinyPerson dataset, while reducing the number of parameters by 23.3%. This significantly enhances the model’s detection performance and robustness, making it highly suitable for object detection tasks on low-computing-power UAV platforms. Full article

Glume-opening of thermosensitive genic male sterile (TGMS) rice (Oryza sativa L.) lines after anthesis is a serious problem that significantly reduces the yield and quality of hybrid seeds. However, the molecular mechanisms regulating the opening and closing of rice glumes remain largely unclear. In this study, we report the isolation and functional characterization of a glum-opening mutant after anthesis, named gom1. gom1 exhibits dysfunctional lodicules that lead to open glumes following anthesis. Map-based cloning and subsequent complementation tests confirmed that GOM1 encodes a receptor-like kinase (RLK). GOM1 was expressed in nearly all floral tissues, with the highest expression in the lodicule. Loss-of-function of GOM1 resulted in a decrease in the expression of genes related to JA biosynthesis, JA signaling, and sugar transport. Compared with LK638S, the JA content in the gom1 mutant was significantly reduced, while the soluble sugar, sucrose, glucose, and fructose contents were significantly increased in lodicules after anthesis. Together, we speculated that GOM1 regulates carbohydrate transport in lodicules during anthesis through JA and JA signaling, maintaining a higher osmolality in lodicules after anthesis, which leads to glum-opening. Full article

Background and Objectives: Computer-aided diagnostic systems have achieved remarkable success in the medical field, particularly in diagnosing malignant tumors, and have done so at a rapid pace. However, the generalizability of the results remains a challenge for researchers and decreases the credibility of these models, which represents a point of criticism by physicians and specialists, especially given the sensitivity of the field. This study proposes a novel model based on deep learning to enhance lung cancer diagnosis quality, understandability, and generalizability. Methods: The proposed approach uses five computed tomography (CT) datasets to assess diversity and heterogeneity. Moreover, the mixup augmentation technique was adopted to facilitate the reliance on salient characteristics by combining features and CT scan labels from datasets to reduce their biases and subjectivity, thus improving the model’s generalization ability and enhancing its robustness. Curriculum learning was used to train the model, starting with simple sets to learn complicated ones quickly. Results: The proposed approach achieved promising results, with an accuracy of 99.38%; precision, specificity, and area under the curve (AUC) of 100%; sensitivity of 98.76%; and F1-score of 99.37%. Additionally, it scored a 00% false positive rate and only a 1.23% false negative rate. An external dataset was used to further validate the proposed method’s effectiveness. The proposed approach achieved optimal results of 100% in all metrics, with 00% false positive and false negative rates. Finally, explainable artificial intelligence (XAI) using Gradient-weighted Class Activation Mapping (Grad-CAM) was employed to better understand the model. Conclusions: This research proposes a robust and interpretable model for lung cancer diagnostics with improved generalizability and validity. Incorporating mixup and curriculum training supported by several datasets underlines its promise for employment as a diagnostic device in the medical industry. Full article

The objective of this article is to create soil burn severity maps to serve as field support for erosion tasks after forest fire occurrence in Spain (2017–2022). The Analytical Spectral Device (ASD) FieldSpec and the CIMEL CE-312 radiometers (optical and thermal, respectively) were used as input data to establish relationships between soil burn severity and reflectance or emissivity, respectively. Spectral indices related to popular forest fire studies and soil assessment were calculated by Sentinel-2 convolved reflectance. All the spectral indices that achieve the separability index algorithm (SI) were validated using specificity, sensitivity, accuracy (ACC), balanced accuracy (BACC), F1-score (F1), and Cohen’s kappa index (k), with 503 field plots. The results displayed the highest overall accuracy results using the Iron Oxide ratio (IOR) index: ACC = 0.71, BACC = 0.76, F1 = 0.63 and k = 0.50, respectively. In addition, IOR was the only spectral index with an acceptable k value (k = 0.50). It is demonstrated that, together with NIR and SWIR spectral bands, the use of blue spectral band reduces atmospheric interferences and improves the accuracy of soil burn severity mapping. The maps obtained in this study could be highly valuable to forest agents for soil erosion restoration tasks. Full article

As one of the important features in the early stage of fires, the detection of smoke can provide a faster early warning of a fire, thus suppressing the spread of the fire in time. However, the features of smoke are not apparent; the shape of smoke is not fixed, and it is easy to be confused with the background outdoors, which leads to difficulties in detecting smoke. Therefore, this study proposes a model called Smoke Detection Transformer (Smoke-DETR) for smoke detection, which is based on a Real-Time Detection Transformer (RT-DETR). Considering the limited computational resources of smoke detection devices, Enhanced Channel-wise Partial Convolution (ECPConv) is introduced to reduce the number of parameters and the amount of computation. This approach improves Partial Convolution (PConv) by using a selection strategy that selects channels containing more information for each convolution, thereby increasing the network’s ability to learn smoke features. To cope with smoke images with inconspicuous features and irregular shapes, the Efficient Multi-Scale Attention (EMA) module is used to strengthen the feature extraction capability of the backbone network. Additionally, in order to overcome the problem of smoke being easily confused with the background, the Multi-Scale Foreground-Focus Fusion Pyramid Network (MFFPN) is designed to strengthen the model’s attention to the foreground of images, which improves the accuracy of detection in situations where smoke is not well differentiated from the background. Experimental results demonstrate that Smoke-DETR has achieved significant improvements in smoke detection. In the self-building dataset, compared to RT-DETR, Smoke-DETR achieves a Precision that has reached 86.2%, marking an increase of 3.6 percentage points. Similarly, Recall has achieved 80%, showing an improvement of 3.6 percentage points. In terms of mAP50, it has reached 86.2%, with a 3.8 percentage point increase. Furthermore, mAP50 has reached 53.9%, representing a 3.6 percentage point increase. Full article

To address the limitations of existing deep learning-based algorithms in detecting surface defects on brake pipe ends, a novel lightweight detection algorithm, FP-YOLOv8, is proposed. This algorithm is developed based on the YOLOv8n framework with the aim of improving accuracy and model lightweight design. First, the C2f_GhostV2 module has been designed to replace the original C2f module. It reduces the model’s parameter count through its unique design. It achieves improved feature representation by adopting specific technique within its structure. Additionally, it incorporates the decoupled fully connected (DFC) attention mechanism, which minimizes information loss during long-range feature transmission by separately capturing pixel information along horizontal and vertical axes via convolution. Second, the Dynamic ATSS label allocation strategy is applied, which dynamically adjusts label assignments by integrating Anchor IoUs and predicted IoUs, effectively reducing the misclassification of high-quality prediction samples as negative samples. Thus, it improves the detection accuracy of the model. Lastly, an asymmetric small-target detection head, FADH, is proposed to utilize depth-separable convolution to accomplish classification and regression tasks, enabling more precise capture of detailed information across scales and improving the detection of small-target defects. The experimental results show that FP-YOLOv8 achieves a mAP50 of 89.5% and an F1-score of 87% on the ends surface defects dataset, representing improvements of 3.3% and 6.0%, respectively, over the YOLOv8n algorithm, Meanwhile, it reduces model parameters and computational costs by 14.3% and 21.0%. Additionally, compared to the baseline model, the AP50 values for cracks, scratches, and flash defects rise by 5.5%, 5.6%, and 2.3%, respectively. These results validate the efficacy of FP-YOLOv8 in enhancing defect detection accuracy, reducing missed detection rates, and decreasing model parameter counts and computational demands, thus meeting the requirements of online defect detection for brake pipe ends surfaces. Full article

Recent advances in cryo-electron microscopy (cryo-EM) have facilitated the high-resolution structural determination of macromolecular complexes in their native states, providing valuable insights into their dynamic behaviors. However, insufficient understanding or experience with the cryo-EM image processing parameters can result in the loss of biological meaning. In this paper, we investigate the dihydrolipoyl acetyltransferase (E2) inner core complex of the pyruvate dehydrogenase complex (PDC) and reconstruct the 3D maps using five different symmetry parameters. The results demonstrate that the reconstructions yield structurally identical 3D models even at a near-atomic structure. This finding underscores a crucial message for researchers engaging in single-particle analysis (SPA) with relatively user-friendly and convenient image processing software. This approach helps reduce the risk of missing critical biological details, such as the dynamic properties of macromolecules. Full article

Background/Objectives: Catha edulis, commonly known as khat, is used for its psychoactive effects and is considered a natural amphetamine. The current study investigated the metabolomic profile in the cerebellum of mice after repeated exposure to khat and evaluated the effects of clavulanic acid on the metabolomic profile in the cerebellum in khat-treated mice. Methods: Male C67BL/6 mice that were 6–9 weeks old were recruited and divided into three groups: the control group was treated with 0.9% normal saline for 17 days; the khat group was given khat extract at a dose of 360 mg/kg via the intraperitoneal (i.p) route for 17 days; and another khat group was treated with khat for 17 days and clavulanic acid at a dose of 5 mg/kg for the last 7 days (days 11–17). At the end of the 17th day, the animals were sacrificed, and their brains were immediately collected and stored at −80 °C. The cerebellum region of the brain was isolated in each group by micropuncture using cryostat and underwent a metabolomics study via Gas Chromatography/Mass Spectroscopy (GC/MS). The total peak area ratios of the selected metabolites in the cerebellum after repeated exposure to the khat extract were significantly reduced (p < 0.05) and treatment of the khat group with clavulanic acid significantly increased (all p < 0.05) the total peak areas ratios of the selected metabolites when compared to their corresponding areas in the alternative khat group. These levels of selected metabolites were further confirmed by observing the metabolite peak area ratios and performing a heat map analysis and a principal compartment analysis of the samples in the cerebellum. Results: A network analysis of altered metabolites in the cerebellum showed a strong correlation between the different metabolites, which showed that an increase in one metabolite can modulate the levels of others. An analysis using the MetaboAnalyst software revealed the involvement of selected altered metabolites like lactic acid in many signaling pathways, like gluconeogenesis, while enrichment analysis data showed altered pathways for pyruvate metabolism and disease pathogenesis. Finally, a network analysis showed that selected metabolites were linked with other metabolites, indicating drug–drug interactions. Conclusions: The present study showed that repeated exposure of mice to khat altered the levels of various metabolites in the cerebellum which are involved in the pathogenesis of different diseases, signaling pathways, and interactions with the pharmacokinetic profile of other therapeutic drugs. The treatment of khat-treated mice with clavulanic acid positively modified the metabolomics profile in the cerebellum and increased the levels of the altered metabolites. Full article

This study introduces BagStacking, an innovative ensemble learning framework designed to enhance the detection of freezing of gait (FOG) in Parkinson’s disease (PD) using accelerometer data. By synergistically combining bagging’s variance reduction with stacking’s sophisticated blending mechanisms, BagStacking achieves superior predictive performance. Evaluated on a comprehensive PD dataset provided by the Michael J. Fox Foundation, BagStacking attained a mean average precision (MAP) of 0.306, surpassing standalone LightGBM and traditional stacking methods. Furthermore, BagStacking demonstrated superior area under the curve (AUC) metrics across key FOG event classes. Specifically, it achieved AUCs of 0.88 for start hesitation, 0.90 for turning, and 0.84 for walking events, outperforming multistrategy ensemble, regular stacking, and LightGBM baselines. Additionally, BagStacking exhibited reduced runtime compared to other ensemble approaches, making it suitable for real-time clinical monitoring. These results underscore BagStacking’s effectiveness in addressing the variability inherent in FOG detection, thereby contributing to improved patient care in PD. Full article