Multi-Level Fusion Net for hand pose estimation in hand-object interaction

This work is about solving a challenging problem of estimating the full 3D hand pose when a hand interacts with an unknown object. Compared to isolated single hand pose estimation, occlusion and interference induced by the manipulated object and the clutter background bring more difficulties for this task. Our proposed Multi-Level Fusion Net focuses on extracting more effective features to overcome these disadvantages by multi-level fusion design with a new end-to-end Convolutional Neural Network (CNN) framework. It takes cropped RGBD data from a single RGBD camera at free viewpoint as input without requiring additional hand–object pre-segmentation and object or hand pre-modeling. Through extensive evaluations on public hand–object interaction dataset, we demonstrate the state-of-the-art performance of our method.     

刮板输送机中部槽耐磨焊条研究

通过对 3种耐磨焊条的金相组织分析 ,X -射线物相分析 ,宏观及显微硬度分析对比 ,选择了Cr-W -B -Nb型耐磨堆焊焊条 ,堆焊层组织为马氏体加残余奥氏体加碳化物 ,其堆焊层硬度高达HRC66。     

安全阀弹簧设计应注意的几个问题

通过对安全阀压力流量特性分析 ,指出了安全阀在弹簧设计时应注意的几个问题     

AZ91D镁合金表面激光熔覆Al+Al2O3涂层研究

为了改善AZ91D镁合金的表面性能,试验利用3kW横流CO2激光器在AZ91D镁合金表面上进行了激光熔覆Al+Al2O3涂层处理。熔覆后使用X射线衍射仪(XRD)和扫描电镜(SEM)对熔覆层进行了微观分析,测试了熔覆层的显微硬度及摩擦磨损性能。试验结果表明:熔覆层由Mg,Mg17Al12,Al2O3等相组成,表层微观结构主要是Mg和Al的亚共晶基体及其上弥散的Al2O3颗粒;熔覆层的平均显微硬度最高达到250HV0.05,明显高于基体AZ91D(70～80HV0.05);其磨损性能与基体相比也有了较大提高。     

基于图像分割技术的压痕直径图像测量

在分析布氏硬度试验压痕图像基础上,提出基于属性直方图的图像分割算法.利用工业CCD相机获取压痕图像,通过图像分割、目标区域处理、压痕圆拟合等步骤完成图像处理,由此实现对压痕尺寸的自动精确测量.实验证明,该算法满足布氏硬度试验对压痕直径的测量要求.     

Power Evaluation of the Los Angeles Oxygen–Fed Ozone System

The Los Angeles Department of Water and Power (LADWP) operates a 600 mgd (2,270 ML/day) direct filtration water treatment plant which includes a 7,900 lb/day (149 kg/hr) ozonation system. Ozone is applied as a preoxidant for the purposes of disinfection and microflocculation. The ozonation system is unique in that high purity oxygen is generated on–site and is used as the feed–gas in a once–through system. The process was selected through competitive bids and evaluated for total present worth from 20–year life cycle costs which included capital plus energy expenses. Power consumption (and penalty) was valued at $6,500/kW. System power demand was measured at nine ozone production rates. A minimum specific energy of 6.5 kWh/lb (14.3 kWh/kg) of ozone was observed when generating ozone at a concentration between 5 and 6 % (wt) [65 and 80 g/m³; where the standard temperature and pressure are 70F (21.11 C) and 1 atm, respectively]. The test methodology and data assessment considerations were developed jointly by the owner and manufacturer, and produced results with practical significance beyond the performance testing objective.     

库尔勒香梨感官及理化指标的定量无损检测

目的 利用高光谱成像技术建立库尔勒香梨分级指标的快速检测方法.方法 选择采摘期香梨作为研究样本,以颜色(a*)、硬度(带皮硬度,hardness)和可溶性固形物(soluble solids content,SSC)为研究指标,使用高光谱成像系统采集样本900～1700 nm范围波长的漫反射光谱.提取样本感兴趣区域(r...     

Fabrication of CuInS2/CNTs absorber layers by sol–gel method

In this work, CuInS₂/multiwalled carbon nanotube (MWCNT) layers are fabricated by the sol–gel spin-coating method. We introduce two forms of MWCNTs into a CIS₂ solution, washed functional multiwalled carbon nanotubes (W-FMWCNTs) and unwashed-functional multiwalled carbon nanotubes (UW-FMWCNTs), in order to investigate the effects of MWCNTs and an acidic environment on the physical properties of the CIS₂ absorber layers. The structure and morphology of the samples are investigated by X-ray diffraction (XRD) and field emission scanning electron microscopy (FESEM), respectively. The XRD study shows that all samples crystallize in a tetragonal structure. The results obtained from the optical, thermo-electric, and electrical measurements indicate that the two groups of CIS₂ layers prepared using W- and UW-FMWCNTs show the opposite behaviors. The Seebeck coefficient (SC) measurements indicate possible formation of a p–n junction.     

Visible–infrared person re-identification based on key-point feature extraction and optimization

Feature extraction for visible–infrared person re-identification (VI-ReID) is challenging because of the cross-modality discrepancy in the images taken by different spectral cameras. Most of the existing VI-ReID methods often ignore the potential relationship between features. In this paper, we intend to transform low-order person features into high-order graph features, and make full use of the hidden information between person features. Therefore, we propose a multi-hop attention graph convolution network (MAGC) to extract robust person joint feature information using residual attention mechanism while reducing the impact of environmental noise. The transfer of higher order graph features within MAGC enables the network to learn the hidden relationship between features. We also introduce the self-attention semantic perception layer (SSPL) which can adaptively select more discriminant features to further promote the transmission of useful information. The experiments on VI-ReID datasets demonstrate its effectiveness.     

In Situ Deformation Topology of COFs with Shortened Channels and High Redox Properties for Li–S Batteries

Covalent organic frameworks (COFs) with various topologies are typically synthesized by selecting and designing connecting units with rich shapes. However, this process is time-consuming and labour-intensive. Besides, the tight stacking of COFs layers greatly restrict their structural advantages. It is crucial to effectively exploit the high porosity and active sites of COFs by topological design. Herein, for the first time, inducing in situ topological changes in sub-chemometric COFs by adding graphene oxide (GO) without replacing the monomer, is proposed. Surprisingly, GO can slow down the intermolecular stacking and induce rearrangement of COFs nanosheets. The channels of D- [4+3] COFs are significantly altered while the stacking of periodically expanded framework is weakened. This not only maximizes the exposure of pore area and polar groups, but also shortens the channels and increases the redox activity, which enables high loading while enhancing host-guest interactions. This topological transformation to exhibit the structural features of COFs for efficient application is an innovative molecular design strategy.