Safety helmet wearing plays a major role in protecting the safety of workers industry and construction, so real-time detection technology is very necessary. This paper proposes an improved YOLOv4 algorithm to achieve efficient detection. The adopts lightweight network PP-LCNet as backbone uses deepwise separable convolution decrease model parameters. Besides, coordinate attention mechanism module embedded three output feature layers enhance information, fusion structure designed fuse target...

Abstract Background This study aims to develop a novel predictive model for determining human papillomavirus (HPV) presence in oropharyngeal cancer using computed tomography (CT). Current image‐based HPV prediction methods are hindered by high computational demands or suboptimal performance. Methods To address these issues, we propose methodology that employs Siamese Neural Network architecture, integrating multi‐modality off‐the‐shelf features—handcrafted features and 3D deep features—to...

Abstract Aiming at the problems of large communication cost consistency process, low efficiency view switching and random selection master node in Practical Byzantine Fault Tolerance (PBFT), an Improved (IPBFT) is proposed. Firstly, This paper proposes a credit model voting mechanism, that generated by with nodes reputation values, which ensures reliability reduces possibility abnormal acts as node. Secondly, process optimized to decrease frequency time between nodes. Finally, protocol...

This study aims to explore the changes of Chinese coronavirus disease-2019 (COVID-19) policy topics in eclipse, outbreak, and convalescent stage COVID-19 based on 4982 textual policies. By using co-occurrence clustering network method, we find that strict prevention control epidemic is only topic policies eclipse stage. In outbreak stage, still most important topic. The resuming work “essential” enterprises stabilizing market prices are support guarantee for fighting against COVID-19. as has...

The traditional centralized logistics resources allocation method can no longer adapt to the new business model of decentralized e-commerce, requiring transaction security for all parties involved in process. Utilizing blockchain and smart contract technologies build network foundation edge computing technology assist resource-constrained transport nodes implementing complex computation, this paper proposes a distributed chain (DLRAChain) concept designs DLRAChain that supports independent...

In the Industry 4.0 era, with continuous integration of industrial field systems and upper-layer facilities, interconnection between wireless sensor networks (IWSNs) Internet is becoming increasingly pivotal. However, when deployed in real scenarios, IWSNs are often connected to legacy control systems, through some wired network protocols via gateways. Complex protocol translation required these gateways, semantic interoperability lacking Internet. To fill this gap, our study focuses on...

Abstract Background and Purpose The immunohistochemical test (IHC) for Human Epidermal Growth Factor Receptor 2 (HER2) hormone receptors (HR) provides prognostic information guides treatment patients with invasive breast cancer. objective of this paper is to establish a non‐invasive system identifying HER2 HR in cancer using dynamic contrast‐enhanced magnetic resonance imaging (DCE‐MRI). Methods In light the absence high‐performance algorithms external validation previously published...

We find that the coherent state projection operator representation of symplectic transformation constitutes a loyal group group. The result successively applying squeezing operators on number can be easily derived.

Abstract Face recognition has become very challenging in unconstrained conditions due to strong intra‐personal variations, such as large pose changes. normalisation can help resolve these problems and effectively improve the face performance by converting non‐frontal faces frontal ones. However, there are other complex facial variations addition pose, illumination expression, which will also influence performance. The authors propose a well‐designed generative adversarial network‐based...

Since traditional satellite positioning technology usually has low accuracy in complex environments, such as indoor environments and dense buildings, wireless can be used a complement to meet the on special scenes. UWB (Ultra Wideband) kind of high-precision technology. However, construction cost system represented by is very expensive. In order requirements reduce cost, we propose an architecture multi-level IoT (Internet Things) this paper. The uses different technologies according...

Fourier and fractional-Fourier transformations are widely used in theoretical physics. In this paper we make quantum perspectives generalization for the fractional transformation (FrFT). By virtue of mechanical representation method integration within normal ordered product (IWOP) operators, find key point composing FrFT, reveal structure FrFT. Following procedure, a full family generalized discovered with usual FrFT as one special case. The eigen-functions arbitrary GFrT derived explicitly.

Based on the Radon transform and fractional Fourier we introduce transformation (FRT). We identify kernel for FRT. The FRT of Wigner operator is derived, which naturally reduces to projector eigenvector rotated quadrature in usual case.

Agile Product Line Engineering (APLE), a relatively new approach combining the two successful methods of Software Development (ASD) and Lines (SPLs), makes product lines more responsive to ever-changing customer needs or market changes. However, SPLs often fail keep up with demand due high coordination costs, slow development processes, long release cycles in case frequent changes business requirements; agile software projects, lack unified specification for describing requirements leads...

Edge-edge collaboration is the main mode for edge computing (EC) in time-sensitive networking (TSN). This distributed mode, which work implemented collaboratively by nodes (ECNs) without involvement of cloud servers, poses challenges to management. However, most existing methods ECN only focus on scheduling resources rather than management process design and rarely consider transparency, security, fair distribution guarantee mechanism benefits. To address this issue, we propose a...

In this paper, we present a joint multi-patch and multi-task convolutional neural networks (JMM-CNNs) framework to learn more descriptive robust face representation for recognition. the proposed JMM-CNNs, set of CNNs feature fusion network are constructed fuse global local facial features, then learning algorithm, including recognition task pose estimation task, is operated on fused obtain pose-invariant task. To further enhance insensitiveness learned representation, also introduce...

For classical Hamiltonian with general form we find a new convenient way to obtain its normal coordinates, namely, let H be quantised and then employ the invariant eigen-operator (IEO) method (Fan et al. 2004 Phys. Lett. A 321 75) derive them. The matrix equation, which relies on M L, for obtaining coordinates of is derived.

Edge computing is a creative paradigm that enhances the capacity of edge device close to data source. As key technology computing, task offloading, which can improve response speed and stability network system, has attracted much attention been applied in many scenarios. However, few studies have considered application offloading time-sensitive networking (TSN), promising potential guarantee delivery with bounded latency low jitter. To this end, we establish stream transmission model for TSN...