Search Results (2,978)

Fire detection is crucial due to the exorbitant annual toll on both human lives and the economy resulting from fire-related incidents. To enhance forest fire detection in complex environments, we propose a new algorithm called FD-Net for various environments. Firstly, to improve detection performance, we introduce a Fire Attention (FA) mechanism that utilizes the position information from feature maps. Secondly, to prevent geometric distortion during image cropping, we propose a Three-Scale Pooling (TSP) module. Lastly, we fine-tune the YOLOv5 network and incorporate a new Fire Fusion (FF) module to enhance the network’s precision in identifying fire targets. Through qualitative and quantitative comparisons, we found that FD-Net outperforms current state-of-the-art algorithms in performance on both fire and fire-and-smoke datasets. This further demonstrates FD-Net’s effectiveness for application in fire detection. Full article

Recently, transformers have demonstrated notable improvements in natural advanced visual tasks. In the field of computer vision, transformer networks are beginning to supplant conventional convolutional neural networks (CNNs) due to their global receptive field and adaptability. Although transformers excel in capturing global features, they lag behind CNNs in handling fine local features, especially when dealing with underwater images containing complex and delicate structures. In order to tackle this challenge, we propose a refined transformer model by improving the feature blocks (dilated transformer block) to more accurately compute attention weights, enhancing the capture of both local and global features. Subsequently, a self-supervised method (a local and global blind-patch network) is embedded in the bottleneck layer, which can aggregate local and global information to enhance detail recovery and improve texture restoration quality. Additionally, we introduce a multi-scale convolutional block attention module (MSCBAM) to connect encoder and decoder features; this module enhances the feature representation of color channels, aiding in the restoration of color information in images. We plan to deploy this deep learning model onto the sensors of underwater robots for real-world underwater image-processing and ocean exploration tasks. Our model is named the refined transformer combined with convolutional block attention module (RT-CBAM). This study compares two traditional methods and six deep learning methods, and our approach achieved the best results in terms of detail processing and color restoration. Full article

Addressing escalating energy demands and greenhouse gas emissions in the oil and gas industry has driven extensive efforts in carbon capture and utilization (CCU), focusing on power plants and industrial facilities. However, utilizing CO₂ as a raw material to produce valuable chemicals, materials, and fuels for transportation may offer a more sustainable and long-term solution than sequestration alone. This approach also presents promising alternatives to traditional chemical feedstock in industries such as fine chemicals, pharmaceuticals, and polymers. This review comprehensively outlines the current state of CO₂ capture technologies, exploring the associated challenges and opportunities regarding their efficiency and economic feasibility. Specifically, it examines the potential of technologies such as chemical looping, membrane separation, and adsorption processes, which are advancing the frontiers of CO₂ capture by enhancing efficiency and reducing costs. Additionally, it explores the various methods of CO₂ utilization, highlighting the potential benefits and applications. These methods hold potential for producing high-value chemicals and materials, offering new pathways for industries to reduce their carbon footprint. The integration of CO₂ capture and utilization is also examined, emphasizing its potential as a cost-effective and efficient approach that mitigates climate change while converting CO₂ into a valuable resource. Finally, the review outlines the challenges in designing, developing, and scaling up CO₂ capture and utilization processes, providing a comprehensive perspective on the technical and economic challenges that need to be addressed. It provides a roadmap for technologies, suggesting that their successful deployment could result in significant environmental benefits and encourage innovation in sustainable practices within the energy and chemical sectors. Full article

During the present study, DNA sequence and morphological data were used to delineate species boundaries in the velvet worm, Peripatopsis sedgwicki species complex. The combined mitochondrial cytochrome c oxidase subunit one (COI) and the nuclear 18S rRNA loci were phylogenetically analyzed using Bayesian inference and maximum likelihood platforms that both demonstrated the presence of four, statistically well-supported clades (A–D). In addition, five species delimitation methods (ASAP, bPTP, bGMYC, STACEY and iBPP) were used on the combined DNA sequence data to identify possible novel lineages. All five species delimitation methods supported the distinction of the Fort Fordyce Nature Reserve specimens in the Eastern Cape province, however, in the main P. sedgwicki s.l. species complex, the species delimitation methods revealed a variable number of novel operational taxonomic units. Gross morphological characters were of limited utility, with only the leg pair number in the Fort Fordyce Nature Reserve specimens and the white head-collar of the Van Stadens Wildflower Nature Reserve specimens being diagnostic. The RADseq results from the earlier study of P. sedgwicki s.l. provided highly congruent results with the four clades observed in the present study. The distribution of P. sedgwicki s.s. (clade B) is restricted to the western portions of its distribution in the Afrotemperate forested regions of the Western Cape Province, South Africa. Three novel species, P. collarium sp. nov., (clade C) P. margaritarius sp. nov., (clade A) and P. orientalis sp. nov., (clade D) are described, of which the first two species are narrow range endemics. The present study, along with several recent systematic studies of velvet worms affirms the importance of fine-scale sampling to detect and document the alpha taxonomic diversity of Onychophora. Full article

The rapid development of generative technologies has made the production of forged products easier, and AI-generated forged images are increasingly difficult to accurately detect, posing serious privacy risks and cognitive obstacles to individuals and society. Therefore, constructing an effective method that can accurately detect and locate forged regions has become an important task. This paper proposes a hierarchical and progressive forged image detection and localization method called HPUNet. This method assigns more reasonable hierarchical multi-level labels to the dataset as supervisory information at different levels, following cognitive laws. Secondly, multiple types of features are extracted from AI-generated images for detection and localization, and the detection and localization results are combined to enhance the task-relevant features. Subsequently, HPUNet expands the obtained image features into four different resolutions and performs detection and localization at different levels in a coarse-to-fine cognitive order. To address the limited feature field of view caused by inconsistent forgery sizes, we employ three sets of densely cross-connected hierarchical networks for sufficient interaction between feature images at different resolutions. Finally, a UNet network with a soft-threshold-constrained feature enhancement module is used to achieve detection and localization at different scales, and the reliance on a progressive mechanism establishes relationships between different branches. We use ACC and F1 as evaluation metrics, and extensive experiments on our method and the baseline methods demonstrate the effectiveness of our approach. Full article

Affected by the complex underwater environment and the limitations of low-resolution sonar image data and small sample sizes, traditional image recognition algorithms have difficulties achieving accurate sonar image recognition. The research builds on YOLOv7 and devises an innovative fast recognition model designed explicitly for sonar images, namely the Dual Attention Mechanism YOLOv7 model (DA-YOLOv7), to tackle such challenges. New modules such as the Omni-Directional Convolution Channel Prior Convolutional Attention Efficient Layer Aggregation Network (OA-ELAN), Spatial Pyramid Pooling Channel Shuffling and Pixel-level Convolution Bilat-eral-branch Transformer (SPPCSPCBiFormer), and Ghost-Shuffle Convolution Enhanced Layer Aggregation Network-High performance (G-ELAN-H) are central to its design, which reduce the computational burden and enhance the accuracy in detecting small targets and capturing local features and crucial information. The study adopts transfer learning to deal with the lack of sonar image samples. By pre-training the large-scale Underwater Acoustic Target Detection Dataset (UATD dataset), DA-YOLOV7 obtains initial weights, fine-tuned on the smaller Smaller Common Sonar Target Detection Dataset (SCTD dataset), thereby reducing the risk of overfitting which is commonly encountered in small datasets. The experimental results on the UATD, the Underwater Optical Target Detection Intelligent Algorithm Competition 2021 Dataset (URPC), and SCTD datasets show that DA-YOLOV7 exhibits outstanding performance, with [email protected] scores reaching 89.4%, 89.9%, and 99.15%, respectively. In addition, the model maintains real-time speed while having superior accuracy and recall rates compared to existing mainstream target recognition models. These findings establish the superiority of DA-YOLOV7 in sonar image analysis tasks. Full article

Cloud infrastructures are designed to provide highly scalable, pay-as-per-use services to meet the performance requirements of users. The workload prediction of the cloud plays a crucial role in proactive auto-scaling and the dynamic management of resources to move toward fine-grained load balancing and job scheduling due to its ability to estimate upcoming workloads. However, due to users’ diverse usage demands, the changing characteristics of workloads have become more and more complex, including not only short-term irregular fluctuation characteristics but also long-term dynamic variations. This prevents existing workload-prediction methods from fully capturing the above characteristics, leading to degradation of prediction accuracy. To deal with the above problems, this paper proposes a framework based on a dual-channel temporal convolutional network and transformer (referred to as DuCFF) to perform workload prediction. Firstly, DuCFF introduces data preprocessing technology to decouple different components implied by workload data and combine the original workload to form new model inputs. Then, in a parallel manner, DuCFF adopts the temporal convolution network (TCN) channel to capture local irregular fluctuations in workload time series and the transformer channel to capture long-term dynamic variations. Finally, the features extracted from the above two channels are further fused, and workload prediction is achieved. The performance of the proposed DuCFF’s was verified on various workload benchmark datasets (i.e., ClarkNet and Google) and compared to its nine competitors. Experimental results show that the proposed DuCFF can achieve average performance improvements of 65.2%, 70%, 64.37%, and 15%, respectively, in terms of Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE) and R-squared ($R^2$) compared to the baseline model CNN-LSTM. Full article

Infrared and visible image fusion can integrate rich edge details and salient infrared targets, resulting in high-quality images suitable for advanced tasks. However, most available algorithms struggle to fully extract detailed features and overlook the interaction of complementary features across different modal images during the feature fusion process. To address this gap, this study presents a novel fusion method based on multi-scale edge enhancement and a joint attention mechanism (MEEAFusion). Initially, convolution kernels of varying scales were utilized to obtain shallow features with multiple receptive fields unique to the source image. Subsequently, a multi-scale gradient residual block (MGRB) was developed to capture the high-level semantic information and low-level edge texture information of the image, enhancing the representation of fine-grained features. Then, the complementary feature between infrared and visible images was defined, and a cross-transfer attention fusion block (CAFB) was devised with joint spatial attention and channel attention to refine the critical supplemental information. This allowed the network to obtain fused features that were rich in both common and complementary information, thus realizing feature interaction and pre-fusion. Lastly, the features were reconstructed to obtain the fused image. Extensive experiments on three benchmark datasets demonstrated that the MEEAFusion proposed in this research has considerable strengths in terms of rich texture details, significant infrared targets, and distinct edge contours, and it achieves superior fusion performance. Full article

Liver cancer is one of the malignancies with high mortality rates worldwide, and its timely detection and accurate diagnosis are crucial for improving patient prognosis. To address the limitations of traditional image segmentation techniques and the U-Net network in capturing fine image features, this study proposes an improved model based on the U-Net architecture, named RHEU-Net. By replacing traditional convolution modules in the encoder and decoder with improved residual modules, the network’s feature extraction capabilities and gradient stability are enhanced. A Hybrid Gated Attention (HGA) module is integrated before the skip connections, enabling the parallel processing of channel and spatial attentions, optimizing the feature fusion strategy, and effectively replenishing image details. A Multi-Scale Feature Enhancement (MSFE) layer is introduced at the bottleneck, utilizing multi-scale feature extraction technology to further enhance the expression of receptive fields and contextual information, improving the overall feature representation effect. Testing on the LiTS2017 dataset demonstrated that RHEU-Net achieved Dice scores of 95.72% for liver segmentation and 70.19% for tumor segmentation. These results validate the effectiveness of RHEU-Net and underscore its potential for clinical application. Full article

This study evaluates the laboratory and field performance of stone mastic asphalt (SMA) mixtures with nominal maximum aggregate sizes (NMAS) of 9.5, 8.0, and 5.6 mm. Aggregates and fine aggregates of these sizes were produced using an impact crusher and a polyurethane screen. Mix designs for SMA overlays on aged concrete pavement were developed. Laboratory tests assessed rutting performance using full-scale accelerated pavement testing (APT) equipment and reflective cracking resistance using an asphalt mixture performance tester (AMPT). Field evaluations included noise reduction using CPX equipment, skid resistance using SN equipment, and bond strength using field cores. Results showed that for 8.0 mm SMA mixtures to achieve the same rutting performance as 9.5 mm SMA, PG76-22 grade binder was required, whereas 5.6 mm SMA required PG82-22. The 8.0 and 5.6 mm SMA mixtures showed 22.2% and 25% reduced crack progression, respectively, compared with the 9.5 mm SMA mixtures. Field evaluations indicated that 8.0 mm and 5.6 mm SMA pavements reduced tire–pavement noise by 1.7 and 0.8 dB, increased skid resistance by 8.5% and 2.0%, and enhanced shear bond strength by 150%, compared with 9.5 mm SMA. Overall, the 8.0 mm SMA mixture on aged concrete pavement demonstrated superior durability and functionality toward sustainable pavement systems. Full article

Accurate forecasting of wind speeds is a crucial aspect of providing fine-scale professional meteorological services (such as wind energy generation and transportation operations etc.). This article utilizes CMA-MESO model forecast data and CARAS-SUR_1 km ground truth grid data from January, April, July, and October 2022, employing the random forest algorithm to establish and evaluate a downscaling correction model for near-surface 1 km resolution wind-speed grid forecast in the complex terrain area of northwestern Hebei Province. The results indicate that after downscaling correction, the spatial distribution of grid forecast wind speeds in the entire complex terrain study area becomes more refined, with spatial resolution improving from 3 km to 1 km, reflecting fine-scale terrain effects. The accuracy of the corrected wind speed forecast significantly improves compared to the original model, with forecast errors showing stability in both time and space. The mean bias decreases from 2.25 m/s to 0.02 m/s, and the root mean square error (RMSE) decreases from 3.26 m/s to 0.52 m/s. Forecast errors caused by complex terrain, forecast lead time, and seasonal factors are significantly reduced. In terms of wind speed categories, the correction significantly improves forecasts for wind speeds below 8 m/s, with RMSE decreasing from 2.02 m/s to 0.59 m/s. For wind speeds above 8 m/s, there is also a good correction effect, with RMSE decreasing from 2.20 m/s to 1.65 m/s. Selecting the analysis of the Zhangjiakou strong wind process on 26 April 2022, it was found that the downscaled corrected forecast wind speed is very close to the observed wind speed at the station and the ground truth grid points. The correction effect is particularly significant in areas affected by strong winds, such as the Bashang Plateau and valleys, which has significant reference value. Full article

To address the challenges of excessive model parameters and low detection accuracy in printed circuit board (PCB) defect detection, this paper proposes a novel PCB defect detection model based on the improved RTDETR (Real-Time Detection, Embedding and Tracking) method, named MFAD-RTDETR. Specifically, the proposed model introduces the designed Detail Feature Retainer (DFR) into the original RTDETR backbone to capture and retain local details. Subsequently, based on the Mamba architecture, the Visual State Space (VSS) module is integrated to enhance global attention while reducing the original quadratic complexity to a linear level. Furthermore, by exploiting the deformable attention mechanism, which dynamically adjusts reference points, the model achieves precise localization of target defects and improves the accuracy of the transformer in complex visual tasks. Meanwhile, a receptive field synthesis mechanism is incorporated to enrich multi-scale semantic information and reduce parameter complexity. In addition, the scheme proposes a novel Multi-frequency Aggregation and Diffusion feature composite paradigm (MFAD-feature composite paradigm), which consists of the Aggregation Diffusion Fusion (ADF) module and the Refiner Feature Composition (RFC) module. It aims to strengthen features with fine-grained awareness while preserving a certain level of global attention. Finally, the Wise IoU (WIoU) dynamic nonmonotonic focusing mechanism is used to reduce competition among high-quality anchor boxes and mitigate the effects of the harmful gradients from low-quality examples, thereby concentrating on anchor boxes of average quality to promote the overall performance of the detector. Extensive experiments are conducted on the PCB defect dataset released by Peking University to validate the effectiveness of the proposed model. The experimental results show that our approach achieves the 97.0% and 51.0% performance in mean Average Precision (mAP)@0.5 and [email protected]:0.95, respectively, which significantly outperforms the original RTDETR. Moreover, the model reduces the number of parameters by approximately 18.2% compared to the original RTDETR. Full article

X-ray absorption fine structure (XAFS) spectroscopy, temperature-programmed reduction (TPR), and temperature-programmed hydride decomposition (TPHD) were employed to elucidate the structures of a series of PdRe/Al₂O₃ bimetallic catalysts for the selective hydrogenation of furfural. TPR evidenced low-temperature Re reduction in the bimetallic catalysts consistent of the migration of [ReO₄]⁻ (perrhenate) species to hydrogen-covered Pd nanoparticles on highly hydroxylated γ-Al₂O₃. TPHD revealed a strong suppression of β-PdH_x formation in the reduced catalysts prepared by (i) co-impregnation and (ii) [HReO₄] impregnation of the reduced Pd/Al₂O₃, indicating the formation of Pd-rich alloy nanoparticles; however, reduced catalysts prepared by (iii) [Pd(NH₃)₄]²⁺ impregnation of calcined Re/Al₂O₃ and subsequent re-calcination did not. Re L_III X-ray absorption edge shifts were used to determine the average Re oxidation states after reduction at 400 °C. XAFS spectroscopy and high-angle annular dark field (HAADF)-scanning transmission electron microscopy (STEM) revealed that a reduced 5 wt.% Re/Al₂O₃ catalyst contained small Re clusters and nanoparticles comprising Re atoms in low positive oxidation states (~1.5+) and incompletely reduced Re species (primarily Re⁴⁺). XAFS spectroscopy of the bimetallic catalysts evidenced Pd-Re bonding consistent with Pd-rich alloy formation. The Pd and Re total first-shell coordination numbers suggest that either Re is segregated to the surface (and Pd to the core) of alloy nanoparticles and/or segregated Pd nanoparticles are larger than Re nanoparticles (or clusters). The Cowley short-range order parameters are strongly positive indicating a high degree of heterogeneity (clustering or segregation of metal atoms) in these bimetallic catalysts. Catalysts prepared using the Pd(NH₃)₄[ReO₄]₂ double complex salt (DCS) exhibit greater Pd-Re intermixing but remain heterogeneous on the atomic scale. Full article

Background/Objectives: Early interventions for infants at high risk of cerebral palsy (CP) are recommended, but limited evidence exists. Our objective was, therefore, to evaluate the effects of the family-centered and interprofessional Small Step early intervention program on motor development in infants at high risk of CP (ClinicalTrials.gov: NCT03264339). Methods: A single-subject research design was employed to investigate participant characteristics (motor dysfunction severity measured using the Hammersmith Infant Neurological Examination (HINE) and Alberta Infant Motor Scale (AIMS) at three months of corrected age (3mCA) related to intervention response. The repeated measures Peabody Developmental Motor Scales-2 fine and gross motor composite (PDMS2-FMC and -GMC) and Hand Assessment for Infants (HAI) were analyzed visually by cumulative line graphs, while the Gross Motor Function Measure-66 (GMFM-66) was plotted against reference percentiles for various Gross Motor Function Classification System (GMFCS) levels. Results: All infants (n = 12) received the Small Step program, and eight completed all five training steps. At two years of corrected age (2yCA), nine children were diagnosed with CP. The children with the lowest HINE < 25 and/or AIMS ≤ 6 at 3mCA (n = 4) showed minor improvements during the program and were classified at GMFCS V 2yCA. Children with HINE = 25–40 (n = 5) improved their fine motor skills during the program, and four children had larger GMFM-66 improvements than expected according to the reference curves but that did not always happen during the mobility training steps. Three children with HINE = 41–50 and AIMS > 7 showed the largest improvements and were not diagnosed with CP 2yCA. Conclusions: Our results indicate that the Small Step program contributed to the children’s motor development, with better results for those with an initial higher HINE (>25). The specificity of training could not be confirmed. Full article

Gold grains are observed in a variety of forms, such as coarse-liberated native gold grains, and ultra-fine grains associated with sulfide or non-sulfide mineral particles, in the form of solid solution in sulfide minerals, mainly pyrite. In the flotation of gold ores, bulk sulfide mineral flotation is generally applied to maximize gold recovery. This approach gives high gold recoveries, but it also causes the recovery of barren sulfide minerals (i.e., sulfide mineral particles with no gold content), which increases concentrate tonnage and transportation costs and reduces the grade sometimes to below the saleable limit (approx. 10 g/t Au). This study addresses the differences between gold-bearing and barren pyrite particles taken from various ore deposits and utilizes these differences for the selective flotation of gold-bearing pyrite. The laboratory scale flotation tests conducted on three pyrite samples having different cyanide soluble gold contents show that a selective separation between gold-bearing pyrite and barren pyrite particles could be achieved under specific flotation conditions. Gold recovery is correlated directly with the cyanide-soluble gold in the ore samples. Electrochemical experiments were conducted to elucidate the differences in surface properties of the two types of pyrite. The barren pyrite particles were more cathodic and prone to cathodic reduction of OH⁻ and depressant ions on the surface, and they could be depressed effectively without significantly affecting the gold-bearing particles. Full article