Objects in unmanned aerial vehicle (UAV) images are generally small due to the high-photography altitude. Although many efforts have been made in object detection, how to accurately and quickly ...detect small objects is still one of the remaining open challenges. In this paper, we propose a feature fusion and scaling-based single shot detector (FS-SSD) for small object detection in the UAV images. The FS-SSD is an enhancement based on FSSD, a variety of the original single shot multibox detector (SSD). We add an extra scaling branch of the deconvolution module with an average pooling operation to form a feature pyramid. The original feature fusion branch is adjusted to be better suited to the small object detection task. The two feature pyramids generated by the deconvolution module and feature fusion module are utilized to make predictions together. In addition to the deep features learned by the FS-SSD, to further improve the detection accuracy, spatial context analysis is proposed to incorporate the object spatial relationships into object redetection. The interclass and intraclass distances between different object instances are computed as a spatial context, which proves effective for multiclass small object detection. Six experiments are conducted on the PASCAL VOC dataset and the two UAV image datasets. The experimental results demonstrate that the proposed method can achieve a comparable detection speed but an accuracy superior to those of the six state-of-the-art methods.
Object tracking in RGB-thermal (RGB-T) videos is increasingly used in many fields due to the all-weather and all-day working capability of the dual-modality imaging system, as well as the rapid ...development of low-cost and miniaturized infrared camera technology. However, it is still very challenging to effectively fuse dual-modality information to build a robust RGB-T tracker. In this paper, an RGB-T object tracking algorithm based on a modal-aware attention network and competitive learning (MaCNet) is proposed, which includes a feature extraction network, modal-aware attention network, and classification network. The feature extraction network adopts the form of a two-stream network to extract features from each modality image. The modal-aware attention network integrates the original data, establishes an attention model that characterizes the importance of different feature layers, and then guides the feature fusion to enhance the information interaction between modalities. The classification network constructs a modality-egoistic loss function through three parallel binary classifiers acting on the RGB branch, the thermal infrared branch, and the fusion branch, respectively. Guided by the training strategy of competitive learning, the entire network is fine-tuned in the direction of the optimal fusion of the dual modalities. Extensive experiments on several publicly available RGB-T datasets show that our tracker has superior performance compared to other latest RGB-T and RGB tracking approaches.
Salvia miltiorrhiza Burge (Danshen) is an eminent medicinal herb that possesses broad cardiovascular and cerebrovascular protective actions and has been used in Asian countries for many centuries. ...Accumulating evidence suggests that Danshen and its components prevent vascular diseases, in particular, atherosclerosis and cardiac diseases, including myocardial infarction, myocardial ischemia/reperfusion injury, arrhythmia, cardiac hypertrophy and cardiac fibrosis. The published literature indicates that lipophilic constituents (tanshinone I, tanshinone IIa, tanshinone IIb, cryptotanshinone, dihydrotanshinone, etc) as well as hydrophilic constituents (danshensu, salvianolic acid A and B, protocatechuic aldehyde, etc) contribute to the cardiovascular protective actions of Danshen, suggesting a potential synergism among these constituents. Herein, we provide a systematic up-to-date review on the cardiovascular actions and therapeutic potential of major pharmacologically active constituents of Danshen. These bioactive compounds will serve as excellent drug candidates in small-molecule cardiovascular drug discovery. This article also provides a scientific rationale for understanding the traditional use of Danshen in cardiovascular therapeutics.
Low impact development (LID) is generally regarded as a more sustainable solution for urban stormwater management than conventional urban drainage systems. However, its effects on urban flooding at a ...scale of urban drainage systems have not been fully understood particularly when different rainfall characteristics are considered. In this paper, using an urbanizing catchment in China as a case study, the effects of three LID techniques (swale, permeable pavement and green roof) on urban flooding are analyzed and compared with the conventional drainage system design. A range of storm events with different rainfall amounts, durations and locations of peak intensity are considered for holistic assessment of the LID techniques. The effects are measured by the total flood volume reduction during a storm event compared to the conventional drainage system design. The results obtained indicate that all three LID scenarios are more effective in flood reduction during heavier and shorter storm events. Their performance, however, varies significantly according to the location of peak intensity. That is, swales perform best during a storm event with an early peak, permeable pavements perform best with a middle peak, and green roofs perform best with a late peak, respectively. The trends of flood reduction can be explained using a newly proposed water balance method, i.e., by comparing the effective storage depth of the LID designs with the accumulative rainfall amounts at the beginning and end of flooding in the conventional drainage system. This paper provides an insight into the performance of LID designs under different rainfall characteristics, which is essential for effective urban flood management.
•Effects of swales, permeable pavements and green roofs on urban flooding are analyzed.•They are more effective in flood reduction during heavier and shorter storm events.•Their performance varies significantly according to the location of storm peak.•The trends of flood reduction can be explained by a proposed water balance method.
Innovative product packaging design is designed to build brand image, increase product sales, and thus enhance economic benefits. This paper explains the principle of the XGBoost algorithm from a ...digital perspective and gives the basic framework. Then the original objective function of XGBoost is approximated and expanded using Taylor’s formula to create the XGBoost classification model, and the performance is evaluated for this model. The XGBoost classification model was used to analyze the data of visual communication appositeness to the production of article shell, including the innovative design of production of article shell illustration and the innovation of visual elements. Among the illustration innovation designs, the illustration designs of life scenes, traditional patterns, famous paintings, and rui animals increased by 44.06%, 42.95%, 66.27%, and 94.25%, respectively. Among the innovation of visual elements, there are six influential elements of product packaging box form, packaging opening method, and packaging material, namely emotional, aesthetic, interesting, interactive, green, and humanistic, accounting for 46.20%, 38.92%, 52.42%, 46.80%, 49.81%, and 42.86% respectively. Promote articles to get rid of traditional thinking packaging production of article shell visual communication under the digital perspective should be based on the reality of life, enhance the design sense of product packaging, and promote the improvement of economic benefits.
Previous studies documented that near‐surface temperatures over the Sahara and Arabian deserts have been amplified in a warming climate, which is termed desert amplification (DA). DA has been linked ...possibly to the large‐scale greenhouse effects associated with increasing water vapour. With very limited moisture availability over the driest desert, two key questions unanswered are the desert moisture sources and the relative contributions of thermodynamic and dynamic processes to the changes in moisture transport. In this study, the atmospheric water vapour budget over the Sahara Desert from 1981–2020 is analysed to address these two questions. Results indicate that the water vapour content over the Sahara Desert has increased significantly since 1981, primarily during the boreal summer and in the lower to middle troposphere. The water vapour budget analysis indicates that in the boreal summer, most of the added moisture is transported into the Sahara Desert through the intensifying northerly inflow across the northern boundary of the desert, while the other boundaries are all export channels. The northerly inward moisture transport is associated with the ridge in the lower troposphere and the Saharan high above the lower troposphere. Further analysis indicates that both dynamic and thermodynamic factors contribute to the increase of the inward moisture flux at the northern boundary, while the associated interannual variability is dominated by the dynamic component related to the circulation pattern changes. The changes of the circulation pattern in the lower troposphere are manifested as the westward extension of the low over the Arabian Peninsula and as the strengthening of the Saharan high above the lower troposphere, both contributing to the increase of the northerly inward moisture transport.
Significant moistening has taken place in the lower to middle troposphere over the Sahara Desert during the boreal summer since 1981. The added moisture is transported into the Sahara Desert across the northern boundary via the intensifying southward moisture flux, while other boundaries are all export channels. Both dynamic and thermodynamic factors contribute to the increase of the northerly inward moisture flux, while the associated interannual variability is dominated by the dynamic component.
On‐skin electrodes function as an ideal platform for collecting high‐quality electrophysiological (EP) signals due to their unique characteristics, such as stretchability, conformal interfaces with ...skin, biocompatibility, and wearable comfort. The past decade has witnessed great advancements in performance optimization and function extension of on‐skin electrodes. With continuous development and great promise for practical applications, on‐skin electrodes are playing an increasingly important role in EP monitoring and human–machine interfaces (HMI). In this review, the latest progress in the development of on‐skin electrodes and their integrated system is summarized. Desirable features of on‐skin electrodes are briefly discussed from the perspective of performances. Then, recent advances in the development of electrode materials, followed by the analysis of strategies and methods to enhance adhesion and breathability of on‐skin electrodes are examined. In addition, representative integrated electrode systems and practical applications of on‐skin electrodes in healthcare monitoring and HMI are introduced in detail. It is concluded with the discussion of key challenges and opportunities for on‐skin electrodes and their integrated systems.
On‐skin electrodes function as an ideal platform for collecting high‐quality electrophysiological signals due to their unique characteristics. This review summarizes the latest progress of on‐skin electrodes, with particular interest in advanced materials for the fabrication, adhesion, and breathability of electrodes, as well as their wearable applications. Future challenges and opportunities for on‐skin electrodes are also discussed.
Nanoporosity of a shale gas reservoir provides essential information on the gas accumulation space and controls the gas reserves. The characteristics of heterogeneous nanoporosity of four shale ...samples are analyzed by combining quantitative evaluation of minerals by scanning electronic microscopy (QEMSCAN), focused ion beam-scanning electron microscopy (FIB-SEM), and nano-CT. The representative elementary area (REA) is proposed by QEMSCAN to detect the imaging area that can represent the overall contents of minerals and organic matter. Combined with the statistics of pores in minerals and organic matter by FIB-SEM, the quantitative nanoporosity is obtained. The nano-CT is used to compare the total nanoporosity that was obtained by FIB-SEM. The results show that shale has distinct characteristics in nanoporosities due to the variation in organic matter and mineral content. The major pore sizes of the organic matter and clay minerals are smaller than 400 nanometers (nm), and the pore sizes of feldspar and pyrite are mainly 200–600 nm. The pore sizes for pores developed in quartz and carbonate minerals range from a few nanometers to 1000 nm. Furthermore, pores smaller than 400 nm mainly provide the total nanoporosity. The nanoporosities in the organic matter are approximately 17%–21%. Since the organic matter content (0.54%–6.98%) is low, the organic matter contributes approximately 5%–33% of the total nanoporosity in shale. Conversely, the nanoporosities in quartz and clay are generally lower than 3%. Since the mineral content (93.02%–99.46%) is obviously higher than the organic matter content, the minerals contribute approximately 67%–95% of the total nanoporosity in shale.
Display omitted
•Heterogeneities of four Longmaxi Formation shale samples were evaluated by representative elementary area.•Nanoporosity is confirmed by the QEMSCAN, FIB-SEM, and Nano-CT methods.•Organic matter with abundant nano-pores contributes about 5%–33% of total nanoporosity.•Nanoporosity is mainly provided by the pores smaller than 400 nm.
To improve the analysis effect of the factors influencing the score of social public management, this paper combines the enhanced regression model to analyze the factors affecting the score of social ...public management and constructs a factor analysis model through an intelligent algorithm. Moreover, this paper uses numerical simulation to verify that when the parameters estimated by SEM are linear functions and the relationship between the research variables and auxiliary variables is nonlinear, the estimation effect of SEM estimators is better than that of traditional generalized regression estimators. In addition, this paper proposes a public service channel management model through the measurement and analysis of channel mode public service feature matching, channel type public service feature matching, and channel type and channel mode public service feature matching. Finally, this paper combines the simulation research to verify the effectiveness of the method in this paper.
Weakly Supervised Video Salient Object Detection (WSVSOD) only requires coarse-grained manual annotations, which can achieve a good trade-off between labeling efficiency and detection performance. In ...this paper, a Multiple Pseudo Label Aggregation Network (MPLA-Net) is proposed for WSVSOD. Firstly, the video frames that can obtain high-quality pseudo labels are selected to generate multiple pseudo labels, so as to avoid the prejudice of the single label. Moreover, the pseudo label with fine edge information is used to generate the Edge Information Map (EIM). Secondly, MPLA-Net is designed to adequately excavate and utilize the comprehensive saliency cues in multiple pseudo labels to improve the detection accuracy, in which ResNet-50 is adopted as the backbone network. Edge loss, pseudo label loss, self-supervised loss and fusion loss are exploited to jointly supervise and optimize the network training to obtain a robust detection model. Experimental results on five benchmark datasets demonstrate that, compared with existing weakly supervised methods, the proposed method can achieve state-of-the-art detection accuracy with less model parameters and higher detection speed. And the detected salient objects have fine boundaries.
PDF
