Search Results (40)

General deep learning methods achieve high-level semantic feature representation by aggregating hierarchical features, which performs well in object detection tasks. However, issues arise with general deep learning methods in UAV-based remote sensing image object detection tasks. Firstly, general feature aggregation methods such as stride convolution may lead to information loss in input samples. Secondly, common FPN methods introduce conflicting information by directly fusing feature maps from different levels. These shortcomings limit the model’s detection performance on small and weak targets in remote sensing images. In response to these concerns, we propose an unmanned aerial vehicle (UAV) object detection algorithm, IF-YOLO. Specifically, our algorithm leverages the Information-Preserving Feature Aggregation (IPFA) module to construct semantic feature representations while preserving the intrinsic features of small objects. Furthermore, to filter out irrelevant information introduced by direct fusion, we introduce the Conflict Information Suppression Feature Fusion Module (CSFM) to improve the feature fusion approach. Additionally, the Fine-Grained Aggregation Feature Pyramid Network (FGAFPN) facilitates interaction between feature maps at different levels, reducing the generation of conflicting information during multi-scale feature fusion. The experimental results on the VisDrone2019 dataset demonstrate that in contrast to the standard YOLOv8-s, our enhanced algorithm achieves a mean average precision (mAP) of 47.3%, with precision and recall rates enhanced by 6.3% and 5.6%, respectively. Full article

This paper presents the parallelization of two widely used implicit numerical solvers for the solution of partial differential equations on structured meshes, namely, the ADI (Alternating-Direction Implicit) solver for tridiagonal linear systems and the SIP (Strongly Implicit Procedure) solver for the penta-diagonal systems. Both solvers were parallelized using CUDA (Computer Unified Device Architecture) Fortran on GPGPUs (General-Purpose Graphics Processing Units). The parallel ADI solver (P-ADI) is based on the Parallel Cyclic Reduction (PCR) algorithm, while the parallel SIP solver (P-SIP) uses the wave front method (WF) following a diagonal line calculation strategy. To map the solution schemes onto the hierarchical block-threads framework of the CUDA on the GPU, the P-ADI solver adopted two mapping methods, one block thread with iterations (OBM-it) and multi-block threads (MBMs), while the P-SIP solver also used two mappings, one conventional mapping using effective WF lines (WF-e) with matrix coefficients and solution variables defined on original computational mesh, and a newly proposed mapping using all WF mesh (WF-all), on which matrix coefficients and solution variables are defined. Both the P-ADI and the P-SIP have been integrated into a two-dimensional (2D) hydrodynamic model, the CCHE2D (Center of Computational Hydroscience and Engineering) model, developed by the National Center for Computational Hydroscience and Engineering at the University of Mississippi. This study for the first time compared these two parallel solvers and their efficiency using examples and applications in complex geometries, which can provide valuable guidance for future uses of these two parallel implicit solvers in computational fluids dynamics (CFD). Both parallel solvers demonstrated higher efficiency than their serial counterparts on the CPU (Central Processing Unit): 3.73~4.98 speedup ratio for flow simulations, and 2.166~3.648 speedup ratio for sediment transport simulations. In general, the P-ADI solver is faster than but not as stable as the P-SIP solver; and for the P-SIP solver, the newly developed mapping method WF-all significantly improved the conventional mapping method WF-e. Full article

“Not In My Back Yard” (NIMBY) conflicts have emerged as a significant challenge in the siting and construction of power grid projects. Traditional risk management methods are often inadequate for addressing the complex interactions between the multiple factors involved in such projects. To explain the relationship between different influencing factors, this paper constructs the hierarchy between the influencing factors using the Interpretive Structural Model (ISM) and carries out a causal analysis of NIMBY conflicts in power grid construction projects using the Bayesian network model. The results of the ISM hierarchical map show that high risk perception and construction practices lacking refinement are the most direct causes of NIMBY incidents. The Bayesian network model indicates that poor construction practices, negative public opinion, high risk perception, inadequate risk assessment and emergency response mechanisms, and negative externalities are the most sensitive factors within the causal system of NIMBY incidents and require priority attention. An increase in risk perception is also found to significantly escalate the severity of NIMBY conflicts. The insights gleaned in this study may provide valuable guidance for managing NIMBY conflicts in power grid construction projects. Full article

Light field (LF) cameras can capture a scene’s information from all different directions and provide comprehensive image information. However, the resulting data processing commonly encounters problems of low contrast and low image quality. In this article, we put forward a content-adaptive light field contrast enhancement scheme using a focal stack (FS) and hierarchical structure. The proposed FS set contained 300 light field images, which were captured using a Lytro-Illum camera. In addition, we integrated the classical Stanford Lytro Light Field Archive and JPEG Pleno Database. Specifically, according to the global brightness, the acquired LF images were classified into four different categories. First, we transformed the original LF FS into a depth map (DMAP) and all-in-focus (AIF) image. The image category was preliminarily determined depending on the brightness information. Then, the adaptive parameters were acquired by the corresponding multilayer perceptron (MLP) network training, which intrinsically enhanced the contrast and adjusted the light field image. Finally, our method automatically produced an enhanced FS based on the DMAP and AIF image. The experimental comparison results demonstrate that the adaptive values predicted by our MLP had high precision and approached the ground truth. Moreover, compared to existing contrast enhancement methods, our method provides a global contrast enhancement, which improves, without over-enhancing, local areas. The complexity of image processing is reduced, and real-time, adaptive LF enhancement is realized. Full article

The concept of symmetry has played a major role in Hilbert space setting owing to the structure of a complete inner product space. Subsequently, different studies pertaining to symmetry, including symmetric operators, have investigated real Hilbert spaces. In this paper, we study the solutions to multiple-set split feasibility problems for a pair of finite families of $\beta$-enriched, strictly pseudocontractive mappings in the setup of a real Hilbert space. In view of this, we constructed an iterative scheme that properly included these two mappings into the formula. Under this iterative scheme, an appropriate condition for the existence of solutions and strong and weak convergent results are presented. No sum condition is imposed on the countably finite family of the iteration parameters in obtaining our results unlike for several other results in this direction. In addition, we prove that a slight modification of our iterative scheme could be applied in studying hierarchical variational inequality problems in a real Hilbert space. Our results improve, extend and generalize several results currently existing in the literature. Full article

In response to the need for precision and intelligence in the assessment of transplanting machine operation quality, this study addresses challenges such as low accuracy and efficiency associated with manual observation and random field sampling for the evaluation of rice seedling planting conditions. Therefore, in order to build a seedling insertion condition detection system, this study proposes an approach based on the combination of image processing and deep learning. The image processing stage is primarily applied to seedling absence detection, utilizing the centroid detection method to obtain precise coordinates of missing seedlings with an accuracy of 93.7%. In the target recognition stage, an improved YOLOv8 Nano network model is introduced, leveraging deep learning algorithms to detect qualified and misplaced seedlings. This model incorporates ASPP (atrous spatial pyramid pooling) to enhance the network’s multiscale feature extraction capabilities, integrates SimAM (Simple, Parameter-free Attention Module) to improve the model’s ability to extract detailed seedling features, and introduces AFPN (Asymptotic Feature Pyramid Network) to facilitate direct interaction between non-adjacent hierarchical levels, thereby enhancing feature fusion efficiency. Experimental results demonstrate that the enhanced YOLOv8n model achieves precision (P), recall (R), and mean average precision (mAP) of 95.5%, 92.7%, and 95.2%, respectively. Compared to the original YOLOv8n model, the enhanced model shows improvements of 3.6%, 0.9%, and 1.7% in P, R, and mAP, respectively. This research provides data support for the efficiency and quality of transplanting machine operations, contributing to the further development and application of unmanned field management in subsequent rice seedling cultivation. Full article

Aquatic vegetation is an essential component of lake ecosystems, used as a biological indicator for in situ monitoring within the Water Framework Directive. We developed a hierarchical object-based image classification model with multi-seasonal Sentinel-2 imagery and suitable spectral indices in order to map the aquatic vegetation in a Mediterranean oligotrophic/mesotrophic deep lake; we then applied the model to another lake with similar abiotic and biotic characteristics. Field data from a survey of aquatic macrophytes, undertaken on the same dates as EO data, were used within the accuracy assessment. The aquatic vegetation was discerned into three classes: emergent, floating, and submerged aquatic vegetation. Geographic object-based image analysis (GEOBIA) proved to be effective in discriminating the three classes in both study areas. Results showed high effectiveness of the classification model in terms of overall accuracy, particularly for the emergent and floating classes. In the case of submerged aquatic vegetation, challenges in their classification prompted us to establish specific criteria for their accurate detection. Overall results showed that GEOBIA based on spectral indices was suitable for mapping aquatic vegetation in oligotrophic/mesotrophic deep lakes. EO data can contribute to large-scale coverage and high-frequency monitoring requirements, being a complementary tool to in situ monitoring. Full article

As a critical foundational component, bearings find widespread application in various mechanical equipment. In order to achieve automated defect detection in the bearing-manufacturing process, a defect detection algorithm combining magnetic particle inspection with deep learning is proposed. Dynamic thresholding and generative adversarial network (GAN) methods are employed to extract defect samples from bearing images and augment the dataset, thereby enhancing data diversity. To mitigate the impact of irrelevant displays in bearing images, a coordinated attention (CA) mechanism is introduced into the backbone network of the deep learning model to focus on key information. Additionally, an adaptive spatial feature fusion module (ASFF) is incorporated during the multiscale fusion stage to maintain consistency in features across different hierarchical levels. The weighted intersection over union (WIoU) bounding box loss function is utilized to replace the original generalized intersection over union (GIoU) in the network, directing the model’s attention towards common-quality anchor boxes to reduce the adverse effects of inconsistent annotations. The experimental results demonstrate that the improved network achieves a mean average precision (mAP) of 98.4% on the bearing dataset, representing a 4.2% improvement over the original network. Full article

Robust evidence suggests that humans explore their environment using a combination of topological landmarks and coarse-grained path integration. This approach relies on identifiable environmental features (topological landmarks) in tandem with estimations of distance and direction (coarse-grained path integration) to construct cognitive maps of the surroundings. This cognitive map is believed to exhibit a hierarchical structure, allowing efficient planning when solving complex navigation tasks. Inspired by human behaviour, this paper presents a scalable hierarchical active inference model for autonomous navigation, exploration, and goal-oriented behaviour. The model uses visual observation and motion perception to combine curiosity-driven exploration with goal-oriented behaviour. Motion is planned using different levels of reasoning, i.e., from context to place to motion. This allows for efficient navigation in new spaces and rapid progress toward a target. By incorporating these human navigational strategies and their hierarchical representation of the environment, this model proposes a new solution for autonomous navigation and exploration. The approach is validated through simulations in a mini-grid environment. Full article

Higher standards have been proposed for detection systems since camouflaged objects are not distinct enough, making it possible to ignore the difference between their background and foreground. In this paper, we present a new framework for Camouflaged Object Detection (COD) named FSANet, which consists mainly of three operations: spatial detail mining (SDM), cross-scale feature combination (CFC), and hierarchical feature aggregation decoder (HFAD). The framework simulates the three-stage detection process of the human visual mechanism when observing a camouflaged scene. Specifically, we have extracted five feature layers using the backbone and divided them into two parts with the second layer as the boundary. The SDM module simulates the human cursory inspection of the camouflaged objects to gather spatial details (such as edge, texture, etc.) and fuses the features to create a cursory impression. The CFC module is used to observe high-level features from various viewing angles and extracts the same features by thoroughly filtering features of various levels. We also design side-join multiplication in the CFC module to avoid detail distortion and use feature element-wise multiplication to filter out noise. Finally, we construct an HFAD module to deeply mine effective features from these two stages, direct the fusion of low-level features using high-level semantic knowledge, and improve the camouflage map using hierarchical cascade technology. Compared to the nineteen deep-learning-based methods in terms of seven widely used metrics, our proposed framework has clear advantages on four public COD datasets, demonstrating the effectiveness and superiority of our model. Full article

Alzheimer’s disease (AD) and cardiovascular traits might share underlying causes. We sought to identify clusters of cardiovascular traits that share genetic factors with AD. We conducted a univariate exome-wide association study and pair-wise pleiotropic analysis focused on AD and 16 cardiovascular traits—6 diseases and 10 cardio-metabolic risk factors—for 188,260 UK biobank participants. Our analysis pinpointed nine genetic markers in the APOE gene region and four loci mapped to the CDK11, OBP2B, TPM1, and SMARCA4 genes, which demonstrated associations with AD at p ≤ 5 × 10⁻⁴ and pleiotropic associations at p ≤ 5 × 10⁻⁸. Using hierarchical cluster analysis, we grouped the phenotypes from these pleiotropic associations into seven clusters. Lipids were divided into three clusters: low-density lipoprotein and total cholesterol, high-density lipoprotein cholesterol, and triglycerides. This split might differentiate the lipid-related mechanisms of AD. The clustering of body mass index (BMI) with weight but not height indicates that weight defines BMI-AD pleiotropy. The remaining two clusters included (i) coronary heart disease and myocardial infarction; and (ii) hypertension, diabetes mellitus (DM), systolic and diastolic blood pressure. We found that all AD protective alleles were associated with larger weight and higher DM risk. Three of the four (75%) clusters of traits, which were significantly correlated with AD, demonstrated antagonistic genetic heterogeneity, characterized by different directions of the genetic associations and trait correlations. Our findings suggest that shared genetic factors between AD and cardiovascular traits mostly affect them in an antagonistic manner. Full article

Some recent studies show that filters in convolutional neural networks (CNNs) have low color selectivity in datasets of natural scenes such as Imagenet. CNNs, bio-inspired by the visual cortex, are characterized by their hierarchical learning structure which appears to gradually transform the representation space. Inspired by the direct connection between the LGN and V4, which allows V4 to handle low-level information closer to the trichromatic input in addition to processed information that comes from V2/V3, we propose the addition of a long skip connection (LSC) between the first and last blocks of the feature extraction stage to allow deeper parts of the network to receive information from shallower layers. This type of connection improves classification accuracy by combining simple-visual and complex-abstract features to create more color-selective ones. We have applied this strategy to classic CNN architectures and quantitatively and qualitatively analyzed the improvement in accuracy while focusing on color selectivity. The results show that, in general, skip connections improve accuracy, but LSC improves it even more and enhances the color selectivity of the original CNN architectures. As a side result, we propose a new color representation procedure for organizing and filtering feature maps, making their visualization more manageable for qualitative color selectivity analysis. Full article

Climate change poses a fundamental threat to the wetlands. The Mediterranean basin is a biodiversity hotspot, and wetlands are important for maintaining this status. The current study evaluated the halophilous vegetation diversity of one of the most relevant Maltese wetlands, Il-Ballut ta’ Marsaxlokk Natura 2000 site, also identified under the Water Framework Directive. A vegetation analysis was carried out according to the Braun–Blanquet approach. The processed dataset included both data from the literature and unpublished data. To quantify vegetation structure and diversity, a hierarchical classification (Chord distance; Ward linkage) and diversity and ecological indices were performed. Diachronic analysis of the taxonomic diversity indices and the Ellenberg indicator values were taken into account. We used an NMDS analysis to assess the ecological fingerprint of the vegetation. In addition, we provided an actual vegetation map for Il-Ballut ta’ Marsaxlokk, based on drone orthophotos. We identified five EU Directive habitats in the study area (1150*, 1310, 1410, 1420, and 92D0) of which one (1150*) was reported for the first time. The ecological fingerprint of the halophilous vegetation has undergone changes over time, particularly due to increasing temperatures. In fact, the results showed that nutrients and temperature were the strongest environmental drivers of the site. The results and methodology of this study demonstrate how vegetation studies can serve as tools to improve knowledge, management actions, and landscape planning of Natura 2000 sites. Full article

In response to problems concerning the low autonomous localization accuracy of mobile robots in unknown environments and large cumulative errors due to long time running, a spatial location representation method incorporating boundary information (SLRB) is proposed, inspired by the mammalian spatial cognitive mechanism. In modeling the firing characteristics of boundary cells to environmental boundary information, we construct vector relationships between the mobile robot and environmental boundaries with direction-aware information and distance-aware information. The self-motion information (direction and velocity) is used as the input to the lateral anti-Hebbian network (LAHN) to generate grid cells. In addition, the boundary cell response values are used to update the grid cell distribution law and to suppress the error response of the place cells, thus reducing the localization error of the mobile robot. Meanwhile, when the mobile robot reaches the boundary cell excitation zone, the activated boundary cells are used to correct the accumulated errors that occur due to long running times, which thus improves the localization accuracy of the system. The main contributions of this paper are as follows: 1. We propose a novel method for constructing boundary cell models. 2. An approach is presented that maps the response values of boundary cells to the input layer of LAHN (Location-Adaptive Hierarchical Network), where grid cells are generated through LAHN learning rules, and the distribution pattern of grid cells is adjusted using the response values of boundary cells. 3. We correct the cumulative error caused by long-term operation of place cells through the activation of boundary cells, ensuring that only one place cell responds to the current location at each individual moment, thereby improving the positioning accuracy of the system. Full article

Information systems are essential instruments in managing resources, in the evaluation of the epidemiological situation, and for decision-making at all hierarchical levels. Technological advances have allowed the development of systems that meet these premises. Therefore, it is recommended to consider the optimization of data entry and its immediate georeferencing in order to obtain information in real time. To meet this objective, we describe the application introduction process for the implementation of the digital collection of primary data and its integration with the database through synchronization with the SisaWeb platform (Information System for surveillance and control of Aedes aegypti), developed to meet the needs of the Arbovirus Surveillance and Control Program in the state of São Paulo, Brazil. For this purpose, the application—SisaMob—was conceived in the Android Studio development environment, Google^®, following the same guidelines as the traditional collection method. Tablets equipped with the Android^® operating system were used. To evaluate the implementation of the application, a semi-structured test was applied. The results highlighted that 774.9% (27) of the interviewees evaluated its use positively and, replacing the standard bulletin, 61.1% (22) of the users considered it regular to excellent. The automatic collection of geographic coordinates represented the greatest innovation in the use of the portable device, with reductions in errors and in the time taken to complete the report in the field. The integration to SisaWeb allowed obtaining information in real-time, being easily presented in tabular and graphic modes and spatially arranged through maps, making it possible to monitor the work at a distance, and allowing preliminary analyses during the data collection process. For the future, we must improve the mechanisms for assessing the effectiveness of information, increase the potential of the tool to produce more accurate analyses, which can direct actions more efficiently. Full article