Single-pixel imaging via compressed sensing can reconstruct high-quality images from a few linear random measurements of an object known a priori to be sparse or compressive, by using a point/bucket ...detector without spatial resolution. Nevertheless, random measurements still have blindness, limiting the sampling ratios and leading to a harsh trade-off between the acquisition time and the spatial resolution. Here, we present a new compressive imaging approach by using a strategy we call cake-cutting, which can optimally reorder the deterministic Hadamard basis. The proposed method is capable of recovering images of large pixel-size with dramatically reduced sampling ratios, realizing super sub-Nyquist sampling and significantly decreasing the acquisition time. Furthermore, such kind of sorting strategy can be easily combined with the structured characteristic of the Hadamard matrix to accelerate the computational process and to simultaneously reduce the memory consumption of the matrix storage. With the help of differential modulation/measurement technology, we demonstrate this method with a single-photon single-pixel camera under the ulta-weak light condition and retrieve clear images through partially obscuring scenes. Thus, this method complements the present single-pixel imaging approaches and can be applied to many fields.
The development of “anode‐free” lithium‐metal batteries with high energy densities is, at present, mainly limited by the poor control of the nucleation of lithium directly on the copper current ...collector, especially in conventional carbonate electrolytes. It is therefore essential to improve the understanding of the lithium nucleation process and its interactions with the copper substrate. In this study, it is shown that diffusion of lithium into the copper substrate, most likely via the grain boundaries, can significantly influence the nucleation process. Such diffusion makes it more difficult to obtain a great number of homogeneously distributed lithium nuclei on the copper surface and thus leads to inhomogeneous electrodeposition. It is, however, demonstrated that the nucleation of lithium on copper is significantly improved if an initial chemical prelithiation of the copper surface is performed. This prelithiation saturates the copper surface with lithium and hence decreases the influence of lithium diffusion via the grain boundaries. In this way, the lithium nucleation can be made to take place more homogenously, especially when a short potentiostatic nucleation pulse that can generate a large number of nuclei on the surface of the copper substrate is applied.
Diffusion of lithium into the copper substrate, most likely via the grain boundaries, can significantly influence the nucleation process by making it more difficult to obtain a great number of homogeneously‐distributed lithium nuclei on the copper surface and thus leads to inhomogeneous electrodeposition. This problem can be mitigated by an initial chemical prelithiation of the copper surface.
3D Convolutional Neural Networks for Human Action Recognition Ji, Shuiwang; Yang, Ming; Yu, Kai
IEEE transactions on pattern analysis and machine intelligence,
2013-Jan., 2013, 2013-Jan, 2013-01-00, 20130101, Letnik:
35, Številka:
1
Journal Article
Recenzirano
We consider the automated recognition of human actions in surveillance videos. Most current methods build classifiers based on complex handcrafted features computed from the raw inputs. Convolutional ...neural networks (CNNs) are a type of deep model that can act directly on the raw inputs. However, such models are currently limited to handling 2D inputs. In this paper, we develop a novel 3D CNN model for action recognition. This model extracts features from both the spatial and the temporal dimensions by performing 3D convolutions, thereby capturing the motion information encoded in multiple adjacent frames. The developed model generates multiple channels of information from the input frames, and the final feature representation combines information from all channels. To further boost the performance, we propose regularizing the outputs with high-level features and combining the predictions of a variety of different models. We apply the developed models to recognize human actions in the real-world environment of airport surveillance videos, and they achieve superior performance in comparison to baseline methods.
This paper examines a stochastic formulation of the generalized Nash equilibrium problem where agents are subject to randomness in the environment of unknown statistical distribution. We focus on ...fully distributed online learning by agents and employ penalized individual cost functions to deal with coupled constraints. Three stochastic gradient strategies are developed with constant step-sizes. We allow the agents to use heterogeneous step-sizes and show that the penalty solution is able to approach the Nash equilibrium in a stable manner within O(μ max ), for small step-size value μ max and sufficiently large penalty parameters. The operation of the algorithm is illustrated by considering the network Cournot competition problem.
Although lithium‐metal electrodes have very high capacities, their use as negative electrodes in batteries is associated with stability and safety problems due to formation of dendrites, mossy as ...well as dead lithium. These problems generally result from the difficulty to ensure that the deposition and stripping of lithium occur homogeneously on the entire electrode surface. As a result, the lithium‐metal electrode is gradually transformed into a thick, porous, and poorly performing electrode. It is therefore essential to develop approaches that facilitate the attainment of homogeneous (i.e., 2D) lithium nucleation and growth. It is also important to note that if the lithium electrode is oxidized on the first half‐cycle, the formed oxidation pits will control the subsequent lithium deposition step. Herein, it is shown that the performance of lithium‐metal electrodes can be straightforwardly improved by introducing a short (e.g., 1 s long) potentiostatic pulse so that the first oxidation step takes place more homogeneously on the lithium surface. This surface activation step gives rise to a large number of preferential lithium nucleation sites facilitating the subsequent attainment of a uniform lithium deposition step. The experimental results indicate that this straightforward pulse approach can significantly increase the lifetime of lithium‐metal electrodes.
A short potentiostatic oxidation/stripping pulse included prior to the first galvanostatic stripping step enables the formation of homogeneously distributed pits which then serve as preferential nucleation sites during the subsequent lithium deposition step. This pulse strategy prevents the deposition from taking place only at the most electrochemically active sites on the lithium‐metal electrodes which improves the stability of the electrodes.
Acute physical activity of moderate to vigorous intensity has been shown to improve cognitive functions in children. However, the empirical evidence associated with Attention Deficit Hyperactivity ...Disorder (ADHD) in children is still limited, in particular regarding which specific cognitive functions benefit. This study investigated the effects of an acute bout of physical activity on multiple aspects of executive functions (inhibition, switching, and visual working memory) in children with ADHD. Forty-six children (8-12 years old; 82.6% boys) were randomly assigned to either 15 minutes of acute exergaming (physical activity of moderate intensity) or to a control condition (sedentary). Executive function performance in inhibition, switching and visual working memory were assessed before and after each condition, using a modified version of both the Flanker and the Color Span Backwards Task. The results revealed that participants in the exergaming group performed significantly faster than those in the control group in terms of both inhibition and switching, but there was no significant difference in the accuracy of the two tasks nor in visual working memory performance. These findings suggest that acute physical activity utilizing exergaming has the potential to improve specific aspects of executive functions (reaction times in inhibition and switching) in children with ADHD.
To summarise the current evidence on the effects of physical activity (PA) interventions on cognitive and academic performance in children, and formulate research priorities and recommendations.
...Systematic review (following PRISMA guidelines) with a methodological quality assessment and an international expert panel. We based the evaluation of the consistency of the scientific evidence on the findings reported in studies rated as of high methodological quality.
PubMed, PsycINFO, Cochrane Central, Web of Science, ERIC, and SPORTDiscus.
PA-intervention studies in children with at least one cognitive or academic performance assessment.
Eleven (19%) of 58 included intervention studies received a high-quality rating for methodological quality: four assessed effects of PA interventions on cognitive performance, six assessed effects on academic performance, and one on both. All high-quality studies contrasted the effects of additional/adapted PA activities with regular curriculum activities. For cognitive performance 10 of 21 (48%) constructs analysed showed statistically significant beneficial intervention effects of PA, while for academic performance, 15 of 25 (60%) analyses found a significant beneficial effect of PA. Across all five studies assessing PA effects on mathematics, beneficial effects were reported in six out of seven (86%) outcomes. Experts put forward 46 research questions. The most pressing research priority cluster concerned the causality of the relationship between PA and cognitive/academic performance. The remaining clusters pertained to PA characteristics, moderators and mechanisms governing the 'PA-performance' relationship and miscellaneous topics.
There is currently inconclusive evidence for the beneficial effects of PA interventions on cognitive and overall academic performance in children. We conclude that there is strong evidence for beneficial effects of PA on maths performance.The expert panel confirmed that more 'high-quality' research is warranted. By prioritising the most important research questions and formulating recommendations we aim to guide researchers in generating high-quality evidence. Our recommendations focus on adequate control groups and sample size, the use of valid and reliable measurement instruments for physical activity and cognitive performance, measurement of compliance and data analysis.
CRD42017082505.
Objective: Myoelectric pattern recognition has been successfully applied as a human-machine interface to control robotic devices such as prostheses and exoskeletons, significantly improving the ...dexterity of myoelectric control. This study investigates the feasibility of applying myoelectric pattern recognition for controlling a robotic hand in stroke patients. Methods: Myoelectric pattern recognition of six hand motion patterns was performed using forearm electromyogram signals in paretic side of eight stroke subjects. Both the random cross validation (RCV) and the chronological handout validation (CHV) were applied to assess the offline myoelectric pattern recognition performance. Experiments on real-time myoelectric pattern recognition control of an exoskeleton robotic hand were also performed. Results: An average classification accuracy of 84.1% (the mean value from two different classifiers) and individual subject differences were observed in the offline myoelectric pattern recognition analysis using the RCV, while the accuracy decreased to 65.7% when the CHV was used. The stroke subjects achieved an average accuracy of 61.3 ± 20.9% for controlling the robotic hand. However, our study did not reveal a clear correlation between the real-time control accuracy and the offline myoelectric pattern recognition performance, or any specific characteristics of the stroke subjects. Conclusion: The findings suggest that it is feasible to apply myoelectric pattern recognition to control the robotic hand in some but not all of the stroke patients. Each stroke subject should be individually online tested for the feasibility of applying myoelectric pattern recognition control for robot-assisted rehabilitation.
The recent development in learning deep representations has demonstrated its wide applications in traditional vision tasks like classification and detection. However, there has been little ...investigation on how we could build up a deep learning framework in a weakly supervised setting. In this paper, we attempt to model deep learning in a weakly supervised learning (multiple instance learning) framework. In our setting, each image follows a dual multi-instance assumption, where its object proposals and possible text annotations can be regarded as two instance sets. We thus design effective systems to exploit the MIL property with deep learning strategies from the two ends; we also try to jointly learn the relationship between object and annotation proposals. We conduct extensive experiments and prove that our weakly supervised deep learning framework not only achieves convincing performance in vision tasks including classification and image annotation, but also extracts reasonable region-keyword pairs with little supervision, on both widely used benchmarks like PASCAL VOC and MIT Indoor Scene 67, and also a dataset for image-and patch-level annotations.
This study examines the effect of online community's recognition on continued participation in idea competitions, and how personal winning record moderates such an influence. We reason that the ...motivating role of community recognition might either be reinforced or substituted by personal winning record, depending upon whether relational motives (psychological and social bonding) or individualistic motives (personal benefits, such as status and career enhancement) are the primary behavior driver. Through an event history analysis of data obtained from a platform of creative design contests, we find that although community recognition exerts a positive effect on the rate of continued participation for designers who are yet to win any competitions, this effect increasingly turns negative for designers who have won. Such findings indicate that the motivating role of community recognition might be substituted instead of reinforced by personal winning record, lending support to the individualistic view while rejecting the relational view. Although virtual social spaces represent an important means for modern competition platforms to attract and motivate participants, our study informs practitioners about online community's limitation in retaining “star” participants.
PDF
