基于CNN的人体姿态识别

姿态识别是人机交互中重要的研究课题之一，随着机器学习与神经网络的发展，研究的方式和成果趋于多样化，姿态识别的应用价值也日趋广泛。本文通过构建卷积神经网络模型，该模型共有11层，在对采样的数据集中5种人体姿态进行卷积与池化操作，最后进入全连接层进行分类，从而完成对数据集的训练和识别。结果显示，相较于机器学习方法，该模型的识别性能更加优秀，且免去了复杂的特征提取方式设计，让网络自身提取特征进行识别分类，效果更好。     

基于结构光的星箭对接环相对位姿测量方法

空间非合作目标位姿测量是空间在轨维护的前提。面向空间机械手对空间载体自动抓捕的应用需求，提出了基于结构光的空间非合作目标视觉测量方法。该方法以双套双线结构光测量装置作为测量传感器，以空间载体上普遍存在的星箭对接环作为抓捕目标。根据对接环的共同特点，选择直线特征和点特征相结合作为相对位姿的求解特征；基于直线特征求解圆环平面法向量，基于点特征求解圆环圆心坐标，进而得到机械手工具坐标系与对接环坐标系间的相对位姿关系；基于多重几何约束实现了不同光照条件下图像上目标直线的鲁棒识别。建立了演示验证实验系统，在大量演示实验的基础上，进行了实验结果分析。     

卷积神经网络用于关节角度识别与姿势评估

为了快速准确地输出各种工作姿势风险评估结果，提出采用Kinect v2与卷积神经网络识别人体各关节角度，并输出标准姿势风险的评估得分。首先使用亚像素角点提取的棋盘标定算法标定Kinect两个摄像头，其次使用改进后的双边滤波对深度图像去噪，使用卷积神经网络识别人体关节二维位置，结合深度信息获取实际三维坐标并计算人体关节角度，最后输出姿势风险评估得分。通过两种实验分别验证了提出的Kinect角度识别与姿势评估的准确性，表明该方法关节角度识别与姿势风险评估的准确率均较高，是一种低成本、高可靠性的姿势评价方法，具有一定的科学意义和工程应用价值。     

Super-Resolution Reconstruction of Images Based on Microarray Camera

In the field of images and imaging, super-resolution (SR) reconstruction of images is a technique that converts one or more low-resolution (LR) images into a highresolution (HR) image. The classical two types of SR methods are mainly based on applying a single image or multiple images captured by a single camera. Microarray camera has the characteristics of small size, multi views, and the possibility of applying to portable devices. It has become a research hotspot in image processing. In this paper, we propose a SR reconstruction of images based on a microarray camera for sharpening and registration processing of array images. The array images are interpolated to obtain a HR image initially followed by a convolution neural network (CNN) procedure for enhancement. The convolution layers of our convolution neural network are 3×3 or 1×1 layers, of which the 1×1 layers are used to improve the network performance particularly. A bottleneck structure is applied to reduce the parameter numbers of the nonlinear mapping and to improve the nonlinear capability of the whole network. Finally, we use a 3×3 deconvolution layer to significantly reduce the number of parameters compared to the deconvolution layer of FSRCNN-s. The experiments show that the proposed method can not only ameliorate effectively the texture quality of the target image based on the array images information, but also further enhance the quality of the initial high resolution image by the improved CNN.     

Learning a Deep Predictive Coding Network for a Semi-Supervised 3D-Hand Pose Estimation

In this paper we present a CNN based approach for a real time 3D-hand pose estimation from the depth sequence. Prior discriminative approaches have achieved remarkable success but are facing two main challenges: Firstly, the methods are fully supervised hence require large numbers of annotated training data to extract the dynamic information from a hand representation. Secondly, unreliable hand detectors based on strong assumptions or a weak detector which often fail in several situations like complex environment and multiple hands. In contrast to these methods, this paper presents an approach that can be considered as semi-supervised by performing predictive coding of image sequences of hand poses in order to capture latent features underlying a given image without supervision. The hand is modelled using a novel latent tree dependency model (LDTM) which transforms internal joint location to an explicit representation. Then the modeled hand topology is integrated with the pose estimator using data dependent method to jointly learn latent variables of the posterior pose appearance and the pose configuration respectively. Finally, an unsupervised error term which is a part of the recurrent architecture ensures smooth estimations of the final pose. Experiments on three challenging public datasets, ICVL, MSRA, and NYU demonstrate the significant performance of the proposed method which is comparable or better than state-of-the-art approaches.      

Enhancing nugget size and weldable current range of ultra-high-strength steel using multi-pulse resistance spot welding

ABSTRACT

This paper presents an innovative approach that uses a pulse-profile to improve the welding quality of CP1180 steel in resistance spot welding process. Three pulses with two cooling times were used in the developed multi-pulse welding (MPW) schedule. The experimental results show that the first pulse increases the contact area between the sheets to improve the current flow pattern. The second pulse was designed to extend the sheet-to-sheet contact area and corona bond for preventing rapid nugget growth. Using these designs, the nugget size was maximised through the third pulse. The maximum nugget size using the designed MPW schedule was 18.5% greater than that of the single-pulse welding schedule and the weldable current range was extended by 130%.     

铁路反恐视频监控系统方案探究

本文通过分析反恐视频监控的需求，研究了普通红外线摄像机和激光摄像机的性能、铁路通信传输网通道容量和视频占用的资源量，推荐反恐视频监控方案采用数字激光摄像机。     

基于深度学习的人体姿态估计方法综述

基于深度学习的人体姿态估计方法旨在通过构建合适的神经网络，直接从二维的图像特征中回归出人体姿态信息。主要按照2D人体姿态估计到3D人体姿态估计的顺序，并从单人检测与多人检测、稀疏的关节点检测与密集的模型构建等方面，对近年来基于深度学习的人体姿态估计方法进行系统介绍，从而初步了解如何通过深度学习的方法得到人体姿态的各个要素，包括肢体部件的相对朝向和比例尺度、骨骼关节点的位置坐标和连接关系，甚至更为复杂的人体蒙皮模型信息。最后,对当前研究面临的挑战以及未来的热点动向进行概述，清晰地呈现出该领域的发展脉络。     

Using the theory of inventive problem solving to brainstorm innovative ideas for assessing varieties of phone-cameras

Today, to survive in an extremely turbulent business environment, traditional products characterized by limited capabilities cannot satisfy diverse customer requirements. In particular, it is observed that the boundaries between smart phones, digital cameras, and tablets are becoming more and more blurred than before. Therefore, for attracting the ad-hoc segments, global companies began to develop hybrid pad-phones and phone-cameras. Unfortunately, some of these products are facing poor sales without incurring much market attention. In order to overcome the aforementioned difficulty, this paper presents a novel framework to reduce the gaps between producer expectation and user perception. By means of the TRIZ (the theory of inventive problem solving), a contradiction matrix is applied to handle engineering conflicts among multi-functional alternatives to seek inventive solutions. Then, ARM (association rule mining) is conducted to identify critical features that formulate customer dissatisfaction (purchase intention). Finally, CA (conjoint analysis) is employed to derive customer utilities for benchmarking varieties of design concepts. In summary, the proposed framework cannot only help product planners efficiently generate innovative ideas, but also effectively justify the validity of design concepts.