Ship recognition method combined with image segmentation and deep learning feature extraction in video surveillance

Multimedia Tools and Applications - To solve the problem of ship recognition in video images, a ship recognition method based on Morphological Watershed image segmentation and Zemike moment is...     

IoT edge computing-enabled collaborative tracking system for manufacturing resources in industrial park

In manufacturing industry, the movement of manufacturing resources in production logistics often affects the overall efficiency. This research is motivated by a world-leading air-conditioner manufacturer. In order to provide the right manufacturing resources for subsequent production steps, excessive time and human effort has been consumed in locating the manufacturing resources in a huge industrial park. The development of Internet of Things (IoT) has made a profound impact on establish smart manufacturing workshop and tracking applications, however a growing trend of data quantity that generated from massive, heterogeneous and bottomed manufacturing resources objects pose challenge to centralized decision. In this study, the concept of edge-computing deeply integrated in collaborative tracking purpose in virtue of IoT technology. An IoT edge computing enabled collaborative tracking architecture is developed to offload the computation pressure and realize distributed decision making. A supervised learning of genetic tracking method is innovatively presented to ensure tracking accuracy and effectiveness. Finally, the research output is developed and implemented in a real-life industrial park for verification. The results show that the proposed tracking method not only performs constant improving accuracy up to 96.14% after learning compared to other tracking method, but also ensure quick responsiveness and scalability.     

Saliency detection analysis of collective physiological responses of pedestrians to evaluate neighborhood built environments

Crowdsourcing pedestrians’ physiological responses (e.g., electrodermal activity (EDA), gait patterns, and blood volume pulse) offers a unique opportunity for assessing and maintaining built environments in a neighborhood. However, raw physiological signals acquired from naturalistic ambulatory settings cannot effectively capture prominent local patterns in the data stream, since diverse technical challenges (e.g., electrode contact noise and motion artifacts) and confounding factors (e.g., heightened physiology due to the movement) make it difficult to detect significant fine-grain signal fluctuations. Motivated by this issue, this paper proposes a method to identify physical disorders that cause pedestrians physical discomfort and/or emotional distress, by using saliency detection analysis on physiological responses. A bottom-up segmentation approach was used as an unsupervised way to divide each physiological signal into homogeneous segments. A physiological saliency cue (PSC) is proposed to calculate the distinctiveness of physiological responses over each segment in contrast to the remaining segments, and the collective PSC of a physical point of interest is computed across participants. The results, obtained from physiological signals collected from wearable devices, indicate that the suggested saliency detection analysis is effectual in capturing prominent local patterns. Our statistical analysis further indicates that the proposed PSC features can be indicative of physical disorders. The outcome of this research will provide a foundation towards using physiological signals to evaluate built environments, and towards promoting neighborhood walkability, increasing feelings of safety in the urban space, and augmenting residents’ well-being.     

A Novel Geometric Mean Feature Space Discriminant Analysis Method for Hyperspectral Image Feature Extraction

Neural Processing Letters - Hyperspectral image contains abundant spectral information with hundreds of spectral continuous bands that allow us to distinguish different classes with more details....     

No-reference image quality assessment based on neighborhood co-occurrence matrix

No-reference image quality assessment (NR-IQA) aims to develop models that can predict the quality of distorted image automatically and accurately in the absent of reference image. Previous NR-IQA methods based on natural scene statistics (NSS) always focus on the luminance contrast of image but attach limited attention to pixel-wise relationship. However, human visual system (HVS) is highly adaptive to extract spatial correlation according to relative position within visual field. In this paper, a new approach is proposed for NR-IQA, in which the neighborhood co-occurrence matrix (NCM) is introduced to describe spatial correlation of pixels for quality assessment. The NCM is constructed based on spatial correlation of every pixel and its neighborhood through a mapping to highlight the one-to-many pixel-wise relationship. Moreover, a series of tailored statistical metrics are designed to quantify the unnaturalness extent of NCM effectively, which is combined with others natural scene statistics to predict image quality. Extensive experiments demonstrate the proposed method has superior performance against compared methods, and achieves significant improvements on distortions associated with color or locality.     

基于区块链的用能数据完整性保护框架

终端设备的分布式拓扑结构和节点有限的计算资源给区块链和用能信息采集系统集成带来很大阻碍,为此,提出基于区块链的多层互连数据保护框架(HBDF),用于能源互联网下的用能数据安全采集.HBDF以分散方式对终端设备进行身份认证来克服单点故障问题以及提高系统的可扩展性.利用基于截断式哈希消息认证码(HMAC)的承诺方案构建轻量级数据完整性验证方案,该方案利用反向哈希链表对验证密钥进行更新来减轻受限节点的计算负担.对框架安全性的分析表明,其可对通信实体和内容进行有效认证.实验验证了该框架的可行性.     

电子货币激励机制综述

电子货币激励机制是信息网络领域普遍采用的方法,在推动资源共享、激发群智感知、促进协作通信等方面有着重要作用,是提升信息网络服务质量与效率的关键.综述电子货币激励机制现有的工作,阐述了电子货币激励机制的挑战,重点介绍了依赖于可信中心的激励机制和基于区块链的分布式激励机制,并探讨了电子货币激励机制中安全可信、隐私保护、可扩展性等问题.     

YAG:Ce3+ nanostructured particles obtained via spray pyrolysis of polymeric precursor solution

Cerium-doped yttrium aluminum garnet (YAG:Ce³⁺) powder phosphor is synthesized via spray pyrolysis of polymeric precursor solution obtained by dissolving the corresponding nitrates in ethylenediaminetetraacetic acid (EDTA). Ultrasonically generated aerosol droplets are decomposed at 600 °C in argon atmosphere. Following the initial attempt in providing pure YAG:Ce³⁺ phase generation the particles were additionally thermally treated for 3 h in air at 1000 and 1100 °C. The powder morphology is followed with scanning electron microscopy (SEM), while inner particle structure is analysed by analytical and high-resolution transmission electron microscopy (TEM). Phase identification is performed by X-ray powder diffraction (XRPD) based on which a structural refinement through Rietveld method was done. The spherical submicronic particles have grained sub-structure comprising clustered garnet monocrystals sized below 100 nm. The YAG:Ce³⁺ emission shows wide peak in the range 470–600 nm with the maximum near 520 nm.     

Investigation of feature contribution to shield tunneling-induced settlement using Shapley additive explanations method

Accurate prediction of shield tunneling-induced settlement is a complex problem that requires consideration of many influential parameters. Recent studies reveal that machine learning(ML) algorithms can predict the settlement caused by tunneling. However, well-performing ML models are usually less interpretable. Irrelevant input features decrease the performance and interpretability of an ML model. Nonetheless, feature selection, a critical step in the ML pipeline, is usually ignored in most stu...