利用机器视觉的检测液晶屏幕坏点分析

本文针对液晶屏幕像素坏点检测当中出现的问题,根据机器视觉原理,利用HALCON以及VS2010作为开发平台,设计了液晶屏幕像素坏点自动检测系统.     

火炮身管结构光检测系统标定方法研究

火炮身管结构光检测系统工作环境视场小、光线弱,导致传统的标定方法存在较大的局限性。针对这一问题,应用结构光成像原理,提出了一种现场标定方法。以155口径火炮为例,设计了内表面包含多种结构信息的标定筒,利用该方法对标定筒内表面结构光图像进行处理,得到了标定筒内凹槽深度与结构光条纹偏移像素距离之间的关系,进而标定了检测系统结构参数。利用标定完成的检测系统对火炮身管膛线的磨损量进行检测,结果表明,检测绝对误差不超过0.01 mm,符合火炮现场检测的要求。     

某型自行火炮水平角速度传感器智能检测仪设计

本项目针对某型自行火炮水平角速度传感器维修需要,以单片机AT89S52为控制单元,设计了自行火炮水平角速度传感器智能检测仪,实现了对水平角速度传感器输出信号状态自动检测与故障判断、信号相位关系自动检测与相位差计算、信号占空比自动检测与辅助调整等功能,为水平角速度传感器的自动快速检测提供有效手段。     

A survey of human pose estimation: The body parts parsing based methods

Estimating human pose from videos and image sequences is not only an important computer vision problem, but also plays very critical role in many real-world applications. Main challenges for human pose estimation are variation of body poses, complicated background and depth ambiguities. To solve these problems, considerable research efforts have been devoted to the related fields. In this survey, we focus our attention on the recent advances in vision-based human pose estimation. We first present a general framework of human pose estimation, and then go through the latest technical progress on each stage. Finally, we discuss the limitations of the existing approaches and foresee the future directions to be explored.     

Real‐Time License Plate Detection in High‐Resolution Videos Using Fastest Available Cascade Classifier and Core Patterns

We present a novel method for real‐time automatic license plate detection in high‐resolution videos. Although there have been extensive studies of license plate detection since the 1970s, the suggested approaches resulting from such studies have difficulties in processing high‐resolution imagery in real‐time. Herein, we propose a novel cascade structure, the fastest classifier available, by rejecting false positives most efficiently. Furthermore, we train the classifier using the core patterns of various types of license plates, improving both the computation load and the accuracy of license plate detection. To show its superiority, our approach is compared with other state‐of‐the‐art approaches. In addition, we collected 20,000 images including license plates from real traffic scenes for comprehensive experiments. The results show that our proposed approach significantly reduces the computational load in comparison to the other state‐of‐the‐art approaches, with comparable performance accuracy.     

Efficient detection of missing tags for passive RFID systems

One of the most interesting radio‐frequency identification applications is the detection of missing tags along with the objects they are attached to. This process must be executed in a fast and efficient way. A monitoring function may need to be periodically executed in order to implement a reliable detection system, therefore raising the need to optimize its performance, in terms of both time and reliability. The contribution of this paper is twofold; on one hand, we provide a protocol taxonomy for the missing‐tag detection problem, which up to our best knowledge is the first of its class, and on the other hand, we propose two novel protocols for a fast detection of the missing tags in a passive radio‐frequency identification system. By increasing the length of the packets exchanged among the reader and the tags, our protocols are able to reduce the time needed to detect the missing tags, by decreasing the number of broadcast messages in the network. Our results show that our proposals outperform the most recent protocols in the literature, obtaining improvements on the execution time going from 25% to 75%. Copyright © 2016 John Wiley & Sons, Ltd.     

Research on detection performance of passive detection system based on troposcatter

Electromagnetic (EM) wave propagated via troposcatter is a valuable candidate for beyond line-of-sight (b-LoS) detection. In this paper, the probability of detection is firstly analyzed. Aiming at acquiring signal-to-noise ratio (SNR), we present two approaches including statistical model and time-varying model. Hopfield model is employed to establish the relationship between SNR and time-varying meteorological parameters. Simulations demonstrate the availability of above models. Finally, a cognitive system is introduced to improve the performance of passive detection system. The uncontrollable factors including meteorological parameters, elevation angle and frequency of EM wave are firstly studied, and then the controllable parameters will be regulated on the basis of conclusions deduced in this paper.     

某火炮故障检测系统设计及精度问题研究

火炮故障检测系统检测范围有限和检测精度低是火炮故障诊断系统的两大缺陷,针对现有火炮检测系统检测精度低的问题,从电路设计、A/D采集电路设计以及液晶显示等方面分析了影响系统检测精度的主要因素,并提出了多种改善检测精度的方案。针对现有火炮检测系统检测范围有限的问题,提出了一个可对炮控箱、操纵台和瞄控箱、电源、开关以及电位旋钮等多位置发生的故障进行检测的火炮综合故障检测系统,并对该系统相关的硬件和软件进行了设计。测试实验表明,该系统功能完善,通用性好,改善系统检测精度的方案实用性强,对于提高我国的火炮故障检测水平具有积极的意义。     

一种光电触发系统的研究

为稳定视觉测量系统中工件在在线自动检测的触装置,以满足相机的触发功能。根据同步光电触发机制原理,分析了触发系统的出射和反射光学光路及其光电探测性能;并设计了相关的信号处理电路,根据电路的噪声特性,分析背景环境对检测电路的影响。通过具体分析计算实验,自动检测功能满足了工件测量系统的触发要求。     

Detecting anomalous spectrum usage in dynamic spectrum access networks

Dynamic spectrum access has been proposed as a means to share scarce radio resources, and requires devices to follow protocols that access spectrum resources in a proper, disciplined manner. For a cognitive radio network to achieve this goal, spectrum policies and the ability to enforce them are necessary. Detection of an unauthorized (anomalous) usage is one of the critical issues in spectrum etiquette enforcement. In this paper, we present a network structure for dynamic spectrum access and formulate the anomalous usage detection problem using statistical significance testing. The detection problem investigated considers two cases, namely, the authorized (primary) transmitter is (i) mobile and (ii) fixed. We propose a detection scheme for each case by exploiting the spatial pattern of received signal energy across a network of sensors. Analytical models are formulated when the distribution of the energy measurements is given and, due to the intractability of the general problem, we present an algorithm using machine learning techniques to solve the general case when the statistics of the energy measurements are unknown. Our simulation results show that our approaches can effectively detect unauthorized spectrum usage with a detection probability above 0.9 while keeping the false alarm rate less than 0.1 when only one unauthorized radio is present, and the detection probability is even higher for more unauthorized radios.     

A survey on packet size optimization for terrestrial,underwater, underground,and body area sensor networks

Packet size optimization is a critical issue in wireless sensor networks (WSNs) for improving many performance metrics (eg, network lifetime, delay, throughput, and reliability). In WSNs, longer packets may experience higher loss rates due to harsh channel conditions. On the other hand, shorter packets may suffer from greater overhead. Hence, the optimal packet size must be chosen to enhance various performance metrics of WSNs. To this end, many approaches have been proposed to determine the optimum packet size in WSNs. In the literature, packet size optimization studies focus on a specific application or deployment environment. However, there is no comprehensive and recent survey paper that categorizes these different approaches. To address this need, in this paper, recent studies and techniques on data packet size optimization for terrestrial WSNs, underwater WSNs, wireless underground sensor networks, and body area sensor networks are reviewed to motivate the research community to further investigate this promising research area. The main objective of this paper is to provide a better understanding of different packet size optimization approaches used in different types of sensor networks and applications as well as introduce open research issues and challenges in this area.     

行人检测技术综述

行人检测是计算机视觉中的研究热点和难点,本文对2005-2011这段时间内的行人检测技术中最核心的两个问题—特征提取、分类器与定位—的研究现状进行综述.文章中首先将这些问题的处理方法分为不同的类别,将行人特征分为底层特征、基于学习的特征和混合特征,分类与定位方法分为滑动窗口法和超越滑动窗口法,并从纵横两个方向对这些方法的优缺点进行分析和比较,然后总结了构建行人检测器在实现细节上的一些经验,最后对行人检测技术的未来进行展望.     

基于粗集信号增强的智能选线实现

随着人们对配电网自动化水平要求的提高,小电流接地故障自动选线问题更加突出,迫切需要从根本上予以解决。因此,研究高准确度、高可靠性的自动选线技术和研制相应自动选线装置,对于提高供电可靠性、减少停电损失和提高配电自动化水平具有重要的意义。本文利用故障暂态特征信息,提出了基于粗集理论的故障选线方法。     

Deep Convolutional Neural Networks for pedestrian detection

Pedestrian detection is a popular research topic due to its paramount importance for a number of applications, especially in the fields of automotive, surveillance and robotics. Despite the significant improvements, pedestrian detection is still an open challenge that calls for more and more accurate algorithms. In the last few years, deep learning and in particular Convolutional Neural Networks emerged as the state of the art in terms of accuracy for a number of computer vision tasks such as image classification, object detection and segmentation, often outperforming the previous gold standards by a large margin. In this paper, we propose a pedestrian detection system based on deep learning, adapting a general-purpose convolutional network to the task at hand. By thoroughly analyzing and optimizing each step of the detection pipeline we propose an architecture that outperforms traditional methods, achieving a task accuracy close to that of state-of-the-art approaches, while requiring a low computational time. Finally, we tested the system on an NVIDIA Jetson TK1, a 192-core platform that is envisioned to be a forerunner computational brain of future self-driving cars.     

Robust people indoor localization with omnidirectional cameras using a Grid of Spatial-Aware Classifiers

This paper describes a system for people indoor localization using omnidirectional cameras and machine learning that significantly reduces the database annotation requirements for the training stage. The most prominent works for people detection are based on machine learning techniques that requires large databases with bounding box annotations (that enclose the people). In this work, a novel multiple classifier system, called Grid of Spatial-Aware Classifiers (GSAC), is proposed, which only requires point-based annotations, allowing to create datasets much faster (and therefore speeding up the system deployment). On the other hand, omnidirectional images have a wider field of view than traditional ones, allowing to monitor a wider area, and thus reducing deployment costs. But in return, they present severe geometric distortions that degrade the performance of state of the art detectors, due to the strong changes in the person appearance caused by the position-dependent distortion. The proposed GSAC satisfactorily addresses this problem by distributing the detection task among all spatial-aware classifiers, so that each classifier has only to deal with a subrange of appearances and distortions. Lastly, a thorough evaluation has been performed on two databases of omnidirectional images: a well-known one and one specifically designed to assess the people detection performance.     

Erasure coding for reliable adaptive retransmission in wireless broadcast/multicast systems

In this paper, we present new adaptive automatic repeat request (ARQ) schemes for wireless broadcast/multicast combining erasure coding (EC) and packet retransmission. Traditional approaches rely on retransmitting the lost packets in a point-to-point or point-to-multipoint mode. The main idea behind the presented protocols is to retransmit adaptive combinations of the lost packets using EC, which can help several receivers to recover the lost information with fewer retransmission attempts. We propose two versions of EC-based ARQ protocols, and investigate theoretically the corresponding transmission bandwidths in different contexts. We show through simulation results the efficiency of the proposed protocols with respect to conventional ARQ strategies and new published ARQ works for broadcast/multicast. Finally, a new sliding window NACK feedback policy is presented for the case of a high number of receivers to avoid the feedback implosion problem.     

智能视频监控中基于机器学习的自动人数统计

提出了智能视频监控中基于机器学习的自动人数统计系统.该系统通过机器学习的方法对视频序列中人的头肩部位进行准确地检测,克服了传统检测方法如连通域分析和简单模板匹配的不足,对光线变化和人群拥挤等问题具有较好的稳健性,在对具体场景的初步测试中取得了较满意的效果.     

Estimator learning automata for feature subset selection in high‐dimensional spaces,case study: Email spam detection

One of the difficult challenges facing data miners is that algorithm performance degrades if the feature space contains redundant or irrelevant features. Therefore, as a critical preprocess task, dimension reduction is used to build a smaller space containing valuable features. There are 2 different approaches for dimension reduction: feature extraction and feature selection, which itself is divided into wrapper and filter approaches. In high‐dimensional spaces, feature extraction and wrapper approaches are not applicable due to the time complexity. On the other hand, the filter approach suffers from inaccuracy. One main reason for this inaccuracy is that the subset's size is not determined considering specifications of the problem. In this paper, we propose ESS (estimator learning automaton‐based subset selection) as a new method for feature selection in high‐dimensional spaces. The innovation of ESS is that it combines wrapper and filter ideas and uses estimator learning automata to efficiently determine a feature subset that leads to a desirable tradeoff between the accuracy and efficiency of the learning algorithm. To find a qualified subset for a special processing algorithm that functions on an arbitrary dataset, ESS uses an automaton to score each candidate subset upon the scale of the subset and accuracy of the learning algorithm using it. In the end, the subset with the highest score is returned. We have used ESS for feature selection in the framework of spam detection, a text classification task for email as a pervasive communication medium. The results show achievement in reaching the goal stated above.     

低层网络攻击及检测方法

常规的探作系统因缺乏充足的审计教据使基于主机的入侵检测系统无法检测到低层网络攻击。基于网络的入侵检测系统因只依靠网上教据流而不能检测到所有攻击。本文分析了几种低层IP攻击，在分析的基础上，提出在探作系统的审计记录中添加部分审计教据，使基于主机的入侵检测系统能检测到低层网络攻击。     

基于STM32的焊缝底片数字化仪硬件设计

针对焊缝缺陷X射线实时自动检测技术普遍存在误检高的问题,研制了焊缝缺陷X射线实时自动检测系统,提出了工业胶片智能检测系统中采集和控制的同步问题的研究方法,设计了步进电机的控制方法与光电编码器采集方法,采用Cortex—M3内核的STM32进行步进电机速度的采集与电机速度PID控制,同时,步进电机带动夹持机构使胶片相对CCD运动,线阵CCD开始采集图像。只要CCD的线频率与扫描机构的运动速度同步,就可以采集到没有畸变的图像,运用LMD18245全桥电机驱动器等器件以及设计所需的相关软件的使用。在此基础上,对系统进行设计、编程和调试,该系统在压力管道焊缝缺陷实时自动检测中验证了其正确性和有效性。