A simple approach to the problem of 3-D reconstruction

A relatively simple mathematical procedure for the reconstruction of the 3-dimensional (3D) image of the left ventricle (LV) of the heart is presented. The method is based on the assumption that every ray whoch emanates from the midpoint of the long axis of the 3D body crosses the surface boundary of the ventricle at one and only one point. The coordinates r_i, φ_i, θ_i of the data points on, say, the outer boundary, (i.e., the epicardium) are calculated in a spherical coordinate system having its origin in the midpoint of the long axis. The problem of defining the coordinates of a prescribed grid point on the boundary is treated as an interpolation problem for the function r = r(φ, θ), defined in the rectangle 0 ≤ φ ≤ 2π; 0 ≤ θ ≤ π with r_i given in the points (φ_i, θ_i).     

Joint relay selection and resource allocation in cooperative device-to-device communications

Device-to-device (D2D) communications allow proximate cellular user equipments (UEs) to communicate with each other directly under the control of base station (BS). In this paper, considering the selection relaying (SR) rule which allows a subset of potential relays to forward the source’s data to the corresponding destination, we first establish a multi-relay system model where a D2D UE can act as not only a source but also a potential relay for another D2D link, and then analyze the cooperation behaviors among selfish UEs. Next a simple strategy is proposed to deal with the relay selection problem, and then the fair and efficient resource sharing problem among cooperating D2D UEs is formulated as a bargaining game. Since the game function is non-convex, we turn to an iterative method by introducing an auxiliary function to get the proportional fair resource allocation results. The system efficiency and fairness are proved by numeral simulation results.     

Large scale automatic image annotation based on convolutional neural network

Automatic image annotation is one of the most important challenges in computer vision, which is critical to many real-world researches and applications. In this paper, we focus on the issue of large scale image annotation with deep learning. Firstly, considering the existing image data, especially the network images, most of the labels of themselves are inaccurate or imprecise. We propose a Multitask Voting (MV) method, which can improve the accuracy of original annotation to a certain extent, thereby enhancing the training effect of the model. Secondly, the MV method can also achieve the adaptive label, whereas most existing methods pre-specify the number of tags to be selected. Additionally, based on convolutional neural network, a large scale image annotation model MVAIACNN is constructed. Finally, we evaluate the performance with experiments on the MIRFlickr25K and NUS-WIDE datasets, and compare with other methods, demonstrating the effectiveness of the MVAIACNN.     

Machine learning to design full-reference image quality assessment algorithm

A crucial step in image compression is the evaluation of its performance, and more precisely, available ways to measure the quality of compressed images. In this paper, a machine learning expert, providing a quality score is proposed. This quality measure is based on a learned classification process in order to respect human observers. The proposed method namely Machine Learning-based Image Quality Measure (MLIQM) first classifies the quality using multi-Support Vector Machine (SVM) classification according to the quality scale recommended by the ITU. This quality scale contains 5 ranks ordered from 1 (the worst quality) to 5 (the best quality). To evaluate the quality of images, a feature vector containing visual attributes describing images content is constructed. Then, a classification process is performed to provide the final quality class of the considered image. Finally, once a quality class is associated to the considered image, a specific SVM regression is performed to score its quality. Obtained results are compared to the one obtained applying classical Full-Reference Image Quality Assessment (FR-IQA) algorithms to judge the efficiency of the proposed method.     

Combining color and spatial information for object recognition across illumination changes

One of the most widely used approaches in the context of object recognition across illumination changes consists in comparing the images by means of the intersection between invariant histograms. However, this approach does not provide satisfying results with limited image databases. We propose to cope with the problem of illumination changes by analyzing simultaneously the color components of the pixels and their spatial arrangement in the image. For this purpose, we introduce the chromatic co-occurrence matrices to characterize the relationship between the color component levels of neighboring pixels. In order to compare two images acquired under different illuminations, these matrices are transformed into adapted co-occurrence matrices that are determined so that their intersection is higher when the two images contain the same object lit with different illuminations than when they contain different objects.     

Thermal conductivity of aligned CNT/polymer composites using mesoscopic simulation

Thermal conductivity of CNT/polymer composites depends on alignment, dispersion, volume fraction and size of CNTs as well as polymer size. By coupling smoothed particle hydrodynamics and dissipative particle dynamics, thermal conductivities of random and aligned composites along with their meso morphologies are studied in detail. Thermal conductivity along the alignment of CNT can be significantly enhanced to 16 times that of polymer by increasing volume fraction, dispersion degree and length of CNT, meanwhile thermal conductivity perpendicular to the alignment of CNT is affected modestly by these factors. Enhancement of thermal conductivity of random composites could only be efficiently achieved by increasing the volume fraction of CNT. Particularly, thermal conductivity $\kappa$ is proportional to the square of volume fraction of CNT v in well dispersed random and aligned composites, i.e. $\kappa \propto v^2$.     

Sketch‐to‐Design: Context‐Based Part Assembly

Designing 3D objects from scratch is difficult, especially when the user intent is fuzzy and lacks a clear target form. We facilitate design by providing reference and inspiration from existing model contexts. We rethink model design as navigating through different possible combinations of part assemblies based on a large collection of pre‐segmented 3D models. We propose an interactive sketch‐to‐design system, where the user sketches prominent features of parts to combine. The sketched strokes are analysed individually, and more importantly, in context with the other parts to generate relevant shape suggestions via adesign galleryinterface. As a modelling session progresses and more parts get selected, contextual cues become increasingly dominant, and the model quickly converges to a final form. As a key enabler, we use pre‐learned part‐based contextual information to allow the user to quickly explore different combinations of parts. Our experiments demonstrate the effectiveness of our approach for efficiently designing new variations from existing shape collections.     

基于伪谱展开的抛物型系统镇定与边界观测研究(英文)

模型截断设计方法在无限维系统控制中得到广泛的应用,但是其自身存在信息丢失的缺陷而限制了控制器对高频扰动抑制的性能.本文研究具有内部和Neumann边界控制的抛物型系统,其中系统采用边界测量.内部控制采用比例反馈形式,其中反馈增益核由Sturm-Liouville系统稳定性分析来待定;类似地,边界反馈的设计也采用待定反馈增益核的方式,最终对描述系统稳定性的Sturm-Liouville系统采用伪谱方法进行求解.数字仿真结果表明了该方法的有效性.     

Power Adjusting Algorithm: A New Cross-Layer Power Saving Mechanism for Mobile Ad-Hoc Networks

Power saving is one of the key issues in Mobile Ad-Hoc Networks(MANETs).It can be realized in Medium Access Control(MAC) layer and network layer.However,previous attentions were mainly paid to MAC layer or network layer with the aim of improving the channel utilization by adopting variable-range transmission power control.In this paper we focus on the power saving in both MAC layer and network layer,and propose a Power Adjusting Algorithm(PAA).In the presence of host’s mobility,PAA is designed to conserve energy by adjusting the transmission power to maintain the route’s connectivity and restarting the route discovery periodically to find a new route with better energy efficiency dynamically.After analyzing the operations of PAA,we find that the length of route discovery restarting period is a critical argument which will affect power saving,and an energy consumption model is abstracted to find the optimal value of the restarting period by analyzing the energy consumption of this algorithm.PAA can handle the mobility of MANET by adjusting the transmission power and in the meantime save energy by restarting route discovery periodically to balance the energy consumption on route discovery and packet delivering.Simulation results show that,PAA saves nearly 40% energy compared with Dynamic Source Routing protocol when the maximum speed of mobile hosts is larger than 8m/s.