The emergence of the cloud storage has brought great convenience to people’s life. Many individuals and enterprises have delivered a large amount data third-party server for storage. Thus, privacy protection retrieved by user needs be guaranteed. Searchable encryption technology environment is adopted ensure that information secure with retrieving data. However, most schemes only support single-keyword search do not file updates, which limit flexibility scheme. To eliminate these problems,...

The Mexican Hat wavelet (MHW) is strictly derived from the heat kernel by taking its negative first-order derivative with respect to time t. As a solution equation that has clear initial condition, Laplace-Beltrami operator. Although MHW descriptor can effectively characterize model information, but it poor robustness scale transformation, and feature description performance affected some extent. Following popular mathematical method, in this paper, we bases on study scaling invariance...

Currently, there is uncertainty in the modeling techniques of cyber-physical systems (CPS) when faced with multiple possibilities and distributions complex system behavior. This leads to system’s inability handle uncertain data correctly, resulting lower reliability model. Additionally, existing technologies struggle verify activity safety CPS after modeling, lacking a dynamic verification analysis approach for properties.This paper introduces generalized possibility decision process as...

An all‐digital PLL (ADPLL) which employs a ΔΣ delay‐locked loop (DLL) to achieve PVT‐insensitive time resolution of the time‐to‐digital converter (TDC) as well noise‐shaped dithering is implemented in 65 nm CMOS. Experimental results show that proposed method can spur reduction with slight degradation in‐band phase noise. The 1.8 GHz ADPLL consumes 14.3 mW, while TDC DLL 2.1 mW.

Both Fuzzy set theory and rough are effective tools in dealing with uncertainty problems. However, these theories have their own inherent limitation, which is the inadequacy of parameterizations tool. Soft a new mathematical tool uncertainties free from above difficulties. can deal broader uncertain problems by combination them. The present paper aims to further generalize intuitionistic fuzzy soft sets investigate application multiattribute decision making. A computing model called...

Abstract The current face-mask recognition detection algorithm during the epidemic only distinguishes between wearing or not a mask. Such often has certain loopholes, such as using other objects to cover their mouths and noses instead of masks cheat detection. To address problems, this paper proposes YOLOv5 based face occlusion algorithm, which is modified on by improving loss function DIoU increasing experimental samples introducing multiple data sets improve object effect. results show...

Remote sensing image captioning has been widely applied to traffic management, geographic research, etc. Although the neural network approach successfully improving performance of system, is still facing object identification challenges due small size objects, uneven distribution, and high coupling with surrounding background. In this paper, we propose a novel remote encoder-decoder model Hierarchical rearrangement-Multi-Layer Perceptron (HMLP) whose encoder adapts hierarchical...

High-quality image generation is an important topic in digital visualization. As a sub-topic of the research, color transfer to produce high-quality with ideal scheme learned from reference one. In this article, we investigate mainstream methods provide survey that introduces related theories and frameworks. Such can be divided into three categories: statistical transfer, semantic-based for special target. For these technical routes, discuss research background, details, representative...

In this paper, the concepts of weighted transducers over strong bimonoids and their input-output-functions are introduced. Further more, input-functions output-functions induced by given. It is most important that can be realized finite automata bimonoids, realization does not depend on distributive law, which also embodies applications bimonoids.

Object detection is a hot talking point in computer vision. Recently, as COVID-19 spreading globally, the epidemic prevention and control has entered normalization, wearing masks when entering leaving public places taking transportation now become normalized. The recognition of face mask also increasing concern. Then fast accurate identification essential. Faster R-CNN currently more advanced object algorithm. It advantages speed high accuracy widely used various fields. However, this method...

With the continuous development of virtual reality, digital image applications, required complex scene video proliferates. For this reason, portrait matting has become a popular topic. In paper, new algorithm with improved details for images backgrounds (MORLIPO) is proposed. This work combines background restoration module (BRM) and fine-grained (FGMatting) to achieve high-detail backgrounds. We recover by inputting single or video, which serves as priori aids in generating more accurate...

Abstract An edge detection technology based on the combination of non-downsampling contour wave transform (NSCT) and tensor voting is proposed, which aims to obtain more accurate detailed information buildings in remote sensing images. Firstly, NSCT used for image decomposition subband frequency different scales angles. Then, position encoding performed these coefficients second-order symmetric tensors at corresponding positions. Tensors angles same are weighted summed complete feature...

Image semantic segmentation algorithm divides the image into several specific regions with unique properties, and extracts objects of interest, it has been widely used in medical analysis, intelligent transportation system, automatic driving other fields. Aiming at problems incomplete boundary segmentation, background interference, easy missed detection, false detection feature extraction today 's algorithms, a HN-DeepLabV3+ is proposed. In encoder, MobileNetV3 lightweight module instead...

The domain of remote sensing image processing has witnessed remarkable advancements in recent years, with deep convolutional neural networks (CNNs) establishing themselves as a prominent approach for building segmentation. Despite the progress, traditional CNNs, which rely on convolution and pooling feature extraction during encoding phase, often fail to precisely delineate global pixel interactions, potentially leading loss vital semantic details. Moreover, conventional CNN-based...