Search Results (6,080)

Automated fish species classification is essential for marine biodiversity monitoring, fisheries management, and ecological research. However, challenges such as environmental variability, class imbalance, and computational demands hinder the development of robust classification models. This study investigates the effectiveness of convolutional neural network (CNN)-based models and hybrid approaches to address these challenges. Eight CNN architectures, including DenseNet121, MobileNetV2, and Xception, were compared alongside traditional classifiers like support vector machines (SVMs) and random forest. DenseNet121 achieved the highest accuracy (90.2%), leveraging its superior feature extraction and generalization capabilities, while MobileNetV2 balanced accuracy (83.57%) with computational efficiency, processing images in 0.07 s, making it ideal for real-time deployment. Advanced preprocessing techniques, such as data augmentation, turbidity simulation, and transfer learning, were employed to enhance dataset robustness and address class imbalance. Hybrid models combining CNNs with traditional classifiers achieved intermediate accuracy with improved interpretability. Optimization techniques, including pruning and quantization, reduced model size by 73.7%, enabling real-time deployment on resource-constrained devices. Grad-CAM visualizations further enhanced interpretability by identifying key image regions influencing predictions. This study highlights the potential of CNN-based models for scalable, interpretable fish species classification, offering actionable insights for sustainable fisheries management and biodiversity conservation. Full article

Automatic glacier extraction from remote sensing images is the most important approach for large scale glacier monitoring. Commonly used band calculation indices to enhance glacier information are not effective in identifying shadowed glaciers and debris-covered glaciers. In this study, we used the Kolmogorov–Smirnov test as the theoretical basis and determined the most suitable band calculation indices to distinguish different land cover classes by comparing inter-sample separability and reasonable threshold range ratios of different indices. We then constructed a glacier classification decision tree. This approach resulted in the development of a method to automatically extract glacier areas at given spatial and temporal scales. In comparison with the commonly used indices, this method demonstrates an improvement in Cohen’s kappa coefficient by more than 3.8%. Notably, the accuracy for shadowed glaciers and debris-covered glaciers, which are prone to misclassification, is substantially enhanced by 108.0% and 6.3%, respectively. By testing the method in the Qilian Mountains, the positive prediction value of glacier extraction was calculated to be 91.8%, the true positive rate was 94.0%, and Cohen’s kappa coefficient was 0.924, making it well suited for glacier extraction. This method can be used for monitoring glacier changes in global mountainous regions, and provide support for climate change research, water resource management, and disaster early warning systems. Full article

The timely and accurate detection of unidentified drones is vital for public safety. However, the unique characteristics of drones in complex environments and the varied postures they may adopt during approach present significant challenges. Additionally, deep learning algorithms often require large models and substantial computational resources, limiting their use on low-capacity platforms. To address these challenges, we propose LAMS-YOLO, a lightweight drone detection method based on linear attention mechanisms and adaptive downsampling. The model’s lightweight design, inspired by CPU optimization, reduces parameters using depthwise separable convolutions and efficient activation functions. A novel linear attention mechanism, incorporating an LSTM-like gating system, enhances semantic extraction efficiency, improving detection performance in complex scenarios. Building on insights from dynamic convolution and multi-scale fusion, a new adaptive downsampling module is developed. This module efficiently compresses features while retaining critical information. Additionally, an improved bounding box loss function is introduced to enhance localization accuracy. Experimental results demonstrate that LAMS-YOLO outperforms YOLOv11n, achieving a 3.89% increase in mAP and a 9.35% reduction in parameters. The model also exhibits strong cross-dataset generalization, striking a balance between accuracy and efficiency. These advancements provide robust technical support for real-time drone monitoring. Full article

This article explores the complex challenges of combating Female Genital Mutilation (FGM) in the counties of Garissa and Isiolo (Kenya) amidst environmental stressors, particularly drought and food insecurity. FGM persists as a deeply entrenched form of violence against women and girls (VAWG), profoundly rooted in social and traditional beliefs. This practice is often justified by cultural norms but leads to severe physical, psychological, and social harm, perpetuating cycles of gender inequality. Framing FGM as VAWG prioritises its recognition as a violation of human rights and underscores the need for comprehensive, culturally sensitive, and rights-based interventions. In addition, this study investigates how these challenges intersect and complicate efforts to address FGM, focusing on two counties in Kenya known for their adherence to this practice. The research methodology employs a longitudinal panel study design, drawing on in-depth narrative interviews with community narrators. A follow-up data collection was planned after six months to monitor changes over time. Furthermore, this research examines the impact of environmental stressors, such as drought and food insecurity, on educational access, resource availability, and community responses to interventions aimed at eradicating FGM. Findings reveal a complex interplay between cultural norms, economic hardships, and the perpetuation of FGM. Drought and food insecurity exacerbate vulnerabilities within communities, diverting attention and resources away from efforts to challenge ingrained practices like FGM. The article underscores the need for holistic approaches that integrate anti-FGM initiatives into broader resilience-building strategies. It emphasises community engagement, culturally sensitive education, and economic empowerment as pivotal elements in addressing FGM within the context of environmental stressors. Recommendations include mobile education units, community-based empowerment programs, and partnerships with local leaders to foster sustainable change. This case study contributes valuable insights into the challenges of combatting FGM in the face of environmental stressors, shedding light on the intricacies that hinder progress. It advocates for a comprehensive, context-specific approach that recognises and navigates the intersecting challenges posed by cultural norms and environmental crises in the fight against FGM. Full article

The integration of artificial intelligence (AI) into project management (PM) transforms how projects are planned, executed, and monitored. The main objective of this study is to provide a comprehensive bibliometric analysis exploring trends, thematic areas, and future directions in AI applications in project management by examining publications from the last decade. This research uncovers dominant themes such as machine learning, decision making, information management, and resource optimization. The findings highlight the growing use of AI to enhance efficiency, accuracy, and innovation in PM processes, with recent trends favoring data-driven approaches and emerging technologies like generative AI. Geographically, China, India, and the United States lead in publications, while the United Kingdom and Australia show a high citation impact. The research landscape, including AI-enhanced decision-making frameworks and cost analysis, demonstrates the diversity of AI applications in PM. An increased interest in the use of generative AI and its impact on PM and project managers was observed. This analysis contributes to the field by offering a structured overview of research trends, defining the challenges and opportunities for integrating AI into project management practices and offering perspectives on emerging technologies. Full article

Laser waveform data that contain rich three-dimensional structural object information hold significant value in forest resource monitoring. However, traditional waveform decomposition algorithms are often constrained by complex waveform structures and depend on the initial parameter selections, which affect the accuracy and robustness of the results. To address the issues of the strong dependence on initial parameters, susceptibility to local optima, and difficulty in detecting hidden peaks during waveform overlap in the traditional satellite laser waveform decomposition algorithms, this study proposes a waveform decomposition method that combines hidden peak detection and an adaptive genetic algorithm (HAGA). This method uses hidden peak detection algorithms to improve the accurate extraction of the Gaussian components from the original waveform and provides the initial parameters. The high-precision extraction of waveform parameters is achieved through the adaptive genetic algorithm (AGA) combined with Levenberg–Marquardt (LM) optimization. In the experimental validation, the proposed method outperformed the traditional methods in both waveform decomposition fitting accuracy and tree height extraction. The average waveform decomposition accuracy $R_{mean}^2$ for more than 2000 laser spots reaches 0.955, whereas the $RMSE$ of the tree height extractions is better than 2 m, demonstrating strong robustness and applicability. Full article

The Psychosocial Rehabilitation Project (PRP) is a tool designed to structure and organize mental health care, guided by the theoretical and practical principles of Psychosocial Rehabilitation (PR). This article aims to identify the initial requirements for the prototyping of a “Psychosocial Rehabilitation Project App”. To achieve this, an integrative review was conducted with the research question: what initial requirements are important to compose the prototype of the “Psychosocial Rehabilitation Project App” in mental health? In the search process, 834 articles were identified and exported to the online systematic review application Rayyan QCRI, resulting in 36 eligible articles for this study, along with one app. The reading of this material allowed the elicitation of three themes: privacy and data protection policy; design; and software and programming. The prototyping of the “Psychosocial Rehabilitation Project App” should prioritize data security and protection, simplicity in design, and the integration of technological resources that facilitate the management, construction, monitoring, and evaluation of psychosocial rehabilitation projects by mental health professionals. Full article

Traditional methods for detecting plant diseases and pests are time-consuming, labor-intensive, and require specialized skills and resources, making them insufficient to meet the demands of modern agricultural development. To address these challenges, deep learning technologies have emerged as a promising solution for the accurate and timely identification of plant diseases and pests, thereby reducing crop losses and optimizing agricultural resource allocation. By leveraging its advantages in image processing, deep learning technology has significantly enhanced the accuracy of plant disease and pest detection and identification. This review provides a comprehensive overview of recent advancements in applying deep learning algorithms to plant disease and pest detection. It begins by outlining the limitations of traditional methods in this domain, followed by a systematic discussion of the latest developments in applying various deep learning techniques—including image classification, object detection, semantic segmentation, and change detection—to plant disease and pest identification. Additionally, this study highlights the role of large-scale pre-trained models and transfer learning in improving detection accuracy and scalability across diverse crop types and environmental conditions. Key challenges, such as enhancing model generalization, addressing small lesion detection, and ensuring the availability of high-quality, diverse training datasets, are critically examined. Emerging opportunities for optimizing pest and disease monitoring through advanced algorithms are also emphasized. Deep learning technology, with its powerful capabilities in data processing and pattern recognition, has become a pivotal tool for promoting sustainable agricultural practices, enhancing productivity, and advancing precision agriculture. Full article

Marine organisms produce sounds for various purposes, including spawning, avoidance, and migration, with each species exhibiting unique acoustic characteristics. This study observed the grunt sounds of Pacific cod (Gadus macrocephalus) during the spawning season for the first time using passive acoustic monitoring (PAM) techniques. Acoustic signals were recorded continuously for about one month at an aquaculture fish farm in Korea. From these recordings, 1208 grunt sounds of Pacific cod were extracted using an automatic grunt detector, and statistical time–frequency parameters were estimated. On average, the grunt sounds consisted of 29 pulses at 6.5 ms intervals within a duration of 205 ms, with a pulse rate of 122.6 per second. The periodic pulse-type signal creates multiple harmonic frequencies on the spectrogram, characterized by time-harmonic modulation with a slope of −240 Hz/s. The mth harmonic frequency distribution ranged from 162 to 822 Hz, with a median source level of 122.6 dB re 1 μPa at 1 m. These findings provide essential scientific data for understanding Pacific cod communication during the spawning season and can aid in identifying spawning sites, conserving habitats, and managing biological resources, contributing to marine ecosystem protection and sustainable management. Full article

Interspecific hybridization can offer advantages in fish aquaculture by enhancing traits like growth rate and disease resistance. However, improper management can result in heterogeneous breeding stocks, which pose risks if hybrids interbreed with native populations. This can lead to loss of genetic diversity and alterations in population structure. This study aimed to evaluate the prevalence of hybridization within the Brazilian aquaculture industry, focusing on the economically significant Serrasalmid species, such as Piaractus mesopotamicus (pacu), Piaractus brachypomus (pirapitinga), Colossoma macropomum (tambaqui), and their hybrids. Using molecular markers (TROP and APOC SNP markers), 312 individuals from the Companhia de Entrepostos e Armazéns Gerais de São Paulo (CEAGESP), the largest warehouse in Latin America, were assessed. The results revealed that 80% of the samples were misidentified, with a substantial proportion of hybrids (72.12%). Among these hybrids, 71.6% were classified as F1 hybrids, while 28.4% were post-F1 hybrids. These findings highlight the need for improved genetic monitoring and management practices in Brazil’s fish production industry, emphasizing the importance of sustainable practices to ensure the long-term viability of aquaculture while preserving native genetic resources. Full article

The Tibetan Plateau, known as the “Third Pole” and the “Water Tower of Asia”, has experienced significant changes in its surface water due to global warming. Accurately understanding and monitoring the spatiotemporal distribution of surface water is crucial for ecological conservation and the sustainable use of water resources. Among existing satellite data, the MODIS sensor stands out for its long time series and high temporal resolution, which make it advantageous for large-scale water body monitoring. However, its spatial resolution limitations hinder detailed monitoring. To address this, the present study proposes a dynamic endmember selection method based on phenological features, combined with mixed pixel decomposition techniques, to generate monthly water abundance maps of the Tibetan Plateau from 2000 to 2023. These maps precisely depict the interannual and seasonal variations in surface water, with an average accuracy of 95.3%. Compared to existing data products, the water abundance maps developed in this study provide better detail of surface water, while also benefiting from higher temporal resolution, enabling effective capture of dynamic water information. The dynamic monitoring of surface water on the Tibetan Plateau shows a year-on-year increase in water area, with an increasing fluctuation range. The surface water abundance products presented in this study not only provide more detailed information for the fine characterization of surface water but also offer a new technical approach and scientific basis for timely and accurate monitoring of surface water changes on the Tibetan Plateau. Full article

The combination of unmanned aerial vehicles and atomic magnetometers can be used for detection applications such as mineral resource exploration, environmental protection, and earthquake monitoring, as well as the detection of sunken ships and unexploded ordnance. A dark-resonance atomic magnetometer offers the significant advantages of a fully optical probe and omnidirectional measurement with no dead zones, making it an ideal choice for airborne applications on unmanned aerial vehicles. Enhancing the sensitivity of such atomic magnetometers is an essential task. In this study, we sought to enhance the sensitivity of a dark-state resonance atomic magnetometer. Initially, through theoretical analysis, we compared the excitation effects of coherent population trapping (CPT) resonance on the D₁ and D₂ transitions of ¹³³Cs thermal vapor. The results indicate that excitation via the D₁ line yields an increase in resonance contrast and a reduction in linewidth when compared with excitation through the D₂ line, aligning with theoretical predictions. Subsequently, considering the impact of various quantum system parameters on sensitivity, as well as their interdependent characteristics, two experimental setups were developed for empirical investigation. One setup focused on parameter optimization experiments, where we compared the linewidth and contrast of CPT resonances excited by both D₁ and D₂ transitions; this led to an optimization of atomic cell size, buffer gas pressure, and operating temperature, resulting in an ideal parameter range. The second setup was employed to validate these optimized parameters using a coupled dark-state atom magnetometer experiment, achieving approximately a 10-fold improvement in sensitivity. Full article

Gaze estimation is increasingly pivotal in applications spanning virtual reality, augmented reality, and driver monitoring systems, necessitating efficient yet accurate models for mobile deployment. Current methodologies often fall short, particularly in mobile settings, due to their extensive computational requirements or reliance on intricate pre-processing. Addressing these limitations, we present Mobile-GazeCapsNet, an innovative gaze estimation framework that harnesses the strengths of capsule networks and integrates them with lightweight architectures such as MobileNet v2, MobileOne, and ResNet-18. This framework not only eliminates the need for facial landmark detection but also significantly enhances real-time operability on mobile devices. Through the innovative use of Self-Attention Routing, GazeCapsNet dynamically allocates computational resources, thereby improving both accuracy and efficiency. Our results demonstrate that GazeCapsNet achieves competitive performance by optimizing capsule networks for gaze estimation through Self-Attention Routing (SAR), which replaces iterative routing with a lightweight attention-based mechanism, improving computational efficiency. Our results show that GazeCapsNet achieves state-of-the-art (SOTA) performance on several benchmark datasets, including ETH-XGaze and Gaze360, achieving a mean angular error (MAE) reduction of up to 15% compared to existing models. Furthermore, the model maintains a real-time processing capability of 20 milliseconds per frame while requiring only 11.7 million parameters, making it exceptionally suitable for real-time applications in resource-constrained environments. These findings not only underscore the efficacy and practicality of GazeCapsNet but also establish a new standard for mobile gaze estimation technologies. Full article

During a widespread epidemic, a large portion of the population faces an increased risk of contracting infectious diseases such as COVID-19, monkeypox, and pneumonia. These outbreaks often trigger cascading effects, significantly impacting society and healthcare systems. To contain the spread, the Centers for Disease Control and Prevention (CDC) must monitor infected individuals (targets) and their geographical locations (areas) as a basis for allocating medical resources. This scenario is a Target-to-Area (TTA) problem. Previous research introduced the Point-In-Polygon (PIP) technique to address multi-target and single-area TTA problems. PIP technology relies on an area’s boundary points to determine whether a target is within that region. However, when dealing with multi-target, multi-area TTA problems, PIP alone may have limitations. The K-Nearest Neighbors (KNN) algorithm presents a promising alternative, but its classification accuracy depends on the availability of sufficient samples, i.e., known targets and their corresponding geographical areas. When sample data are limited, the effectiveness of KNN is constrained, potentially delaying the CDC’s ability to track and manage outbreaks. For this problem, this study proposes an improved approach that integrates PIP and KNN technologies while introducing area boundary points as additional samples. This enhancement aims to improve classification accuracy and mitigate the impact of insufficient sample data on epidemic tracking and management. Full article

Quantifying the dynamic changes and relationships between ecosystem services (ESs) and land use change is critical for sustainable ecosystem management and land use optimization. However, comprehensive discussions on the spatiotemporal variations of ESs and their relationships with land use intensity (LUI) are lacking, particularly in the context of significant climate warming. Systematic analyses of the forestry management unit scale are limited, leading to considerable uncertainty in sustainable ecosystem management, especially in alpine ecosystems of the Northern Hemisphere, where ESs have significantly degraded. The study focuses on the Daxing’anling forest area, Inner Mongolia (DFIAM), a representative sensitive alpine ecosystem and crucial ecological security barrier in Northern China. Utilizing the InVEST model, we analyzed the spatiotemporal variations in land use and four essential ESs, water yield (WY), carbon storage (CS), soil conservation (SC), and habitat quality (HQ), from 2013 to 2018. We also assessed the dynamic relationships between LUI and these ESs using a four-quadrant model. Our findings indicate the following: (1) Land use types in DFIAM remained relatively stable between 2013 and 2018, with forest being the dominant type (approximately 93%). During this period, areas of forest, cropland, impervious surfaces, and bare land increased, while areas of grassland, water, and wetland decreased. Although the overall change of LUI was gentle, a spatial pattern of “high in the southeast and low in the northwest” emerged, with low LUI areas showing slight expansion. (2) WY, SC, and HQ decreased, while CS increased from 2013 to 2018. The spatial distributions of these ESs showed higher values in the center and lower values at the edges, with forests demonstrating a strong capacity to provide multiple ESs. (3) The relationship between LUI and the four ESs from 2013 to 2018 was predominantly negative, primarily situated in Quadrant II, indicating that increased LUI inhibited ES supply capacity. Within Quadrant II, the distribution range of LUI, WY, and HQ decreased, while CS remained stable and SC increased. Furthermore, Quadrant III (positive correlation) accounted for a large proportion (19.23%~42.31%), highlighting the important role of non-anthropogenic factors in ES changes. Overall, most ESs in the DFAIM showed a decline while LUI remained relatively stable, with predominantly negative correlations between LUI and ESs. The increased LUI driven by human activities, and other non-human factors, may have contributed significantly to ES degradation. To improve ESs, we proposed implementing differentiated land use planning and management, systematic ecological protection and restoration strategies, a multi-level ecological early-warning monitoring and evaluation network, ecological corridors and buffer zones, and a collaborative management system with multiple participation. These results provide scientific guidance for the sustainable management of alpine ecosystems, enhancement of ESs, and formulation of land resource protection policies. Full article