Traffic accidents caused by distracted drivers account for a large proportion of traffic each year, and monitoring the driving state to avoid has become very important research direction. At present, field driver distraction detection mainly adopts supervised learning methods, which have problems such as poor generalization ability, labeling cost, weak artificial intelligence. This paper is oriented toward fine-grained innovatively proposes new unsupervised deep algorithm, referred UDL,...

In this work, we propose a voxel-based single-stage fine-grained and efficient point cloud 3D object detection algorithm to address the inadequate granularity in feature extraction tasks imbalance between efficiency accuracy scenarios. We develop lightweight multibranch cross-sparse convolution network (LMCCN) that is designed preserve of original while achieving enhanced efficiency. Additionally, introduce compact self-attention augmented bird's eye view (BEV) module (CFSAM). This aims...

According to the surveys of World Health Organization, distracted driving is one main causes road traffic accidents. To improve safety, real-time detection drivers' behavior very important for development highly reliable Advanced Driver Assistance System (ADAS). At present, deep learning architecture based on a Convolutional Neural Network (CNN) has disadvantages such as large number parameters and weak global feature extraction ability. Therefore, this paper proposes an innovative driver...

Effective and efficient retargeting techniques may enrich users' browsing experiences in mobile devices. Existing mesh-based solutions put less efforts making well-tuned meshes. In this paper, we propose a novel adaptive grid based optimization method to retarget an image. First, present entropy measure guide the construction. Then employ quadtree structure adjust granularity adaptively. Furthermore, reduce inappropriate deformation from inconsistent importance assignment, build global model...

Imaging in the natural scene under ill lighting conditions (e.g., low light, back-lit, over-exposed front-lit, and any combinations of them) suffers from both over- under-exposure at same time, whereas processing such images often results under-enhancement. A single small image sensor can hardly provide satisfactory quality for with ordinary optical lenses capturing devices. Challenges arise maintenance a visual smoothness between those regions, while color contrast should be well preserved....

The popularity of stereo images and various sizes display screens pose the need image retargeting techniques which resize pairs to desired sizes. Many content-aware methods adapt through non-uniformly resizing regions. However, these often make depth retargeted version inconsistent with original one, since they do not explicitly consider different effects distinct regions on depths 3D scenes. In this paper, we analyze region-wise With such insights, can properly edit or maintain a pair via...

Take the perspective of cool hue and warm hue, this paper investigated atmosphere perception LED dynamic colored lighting, as well physical parameters effects. An experiment was carried out in a lighting lab with 8 fully functioned cubes, which can produce temporal color-varied lighting. 30 Chinese observers (14 males, 16 females, mean age=25.43, SD= 2.86) evaluated 36 Light conditions (2 lightness levels × 2 chroma 3 range types speed) 25 word pairs questionnaire. 4 dimensions were...

Abstract Integrating renewable energy sources like wind power into the grid enhances dynamic interactions among energy-producing equipment, leading to new technological issues for grid. Modeling and simulation are essential ensure stability of emerging grid, but they require precise component modeling, which is often unavailable due technical confidentiality other factors. Conventional hardware-in-the-loop (HIL) can accurately simulate dynamics a single device, not complex multiple devices....

Abstract The morphological structure and dynamic characteristics of power distribution systems are rapidly evolving due to the widespread application distributed renewable energy rapid advancement electronics technology. stability analysis these complex new depends on electromagnetic transient (EMT) simulation. However, intellectual property rights protection by manufacturers leads many network devices that can only use black-box models with missing parameters, which challenges This reduces...

In the era of e-business, traditional business services are greatly challenged by ever-increasing demands from customers with various backgrounds and personalities. Large numbers new e-businesses driven needs customers. This paper proposes a situation calculus based approach to dynamically manage e-business in ubiquitous networked environment. By employing this formalism, can implement service model verification automatic composition.

An arbiter physical unclonable function (APUF) has exponential challenge-response pairs and is easy to implement on field-programmable gate arrays (FPGAs). However, modeling attacks based machine learning have become a serious threat APUFs. Although the modeling-attack resistance of an MA-APUF been improved considerably by architecture modifications, response generation method results in low uniqueness. In this study, we demonstrate three design problems regarding uniqueness that APUF-based...

This paper describes a W-band microstrip reflectarray antenna with single-layer substrate. The reflector element is double resonant ring-round mixed structure, in which the circle radius and ring inner are changed. combination of two types elements can achieve full 360° range phase shifts. Using this technique, 50 by has been designed. feasibility design verified simulated measured results.

This paper presents a 3D perceptual system based on 2D laser radar LMS291. The proposed utilizes intelligent module to add degree of freedom A new method synchronization between LMS291 and was proposed. We added Visual C++ multi-media timer the timestamp method, purpose we use this is change level resolution conveniently, control number points acquired, because in some condition need speedy scanning rebuild reconstructed image terrain quickly. And Matlab reconstruct images. To repair...

For given a single input and output controllability or observability linear control system, one can transfer it to canonical form. In this paper the authors consider an inverse problem of above theory. That is, for we construct new polynomial system which is algebraicly equivalent one. Its observability/controllability matrices as well associated generalized Bezout matrix are studied. Meanwhile, some analogous expression formulas in terms obtained.