Practical design and implementation of recognition assisted dynamic surveillance system

Open Service Gateway Initiative (OSGi) and Open-source Computer Vision (OpenCV) are widely used for developing applications. OSGi is constructed to provide a service platform with high application interoperability while OpenCV is used to provide many application programming interfaces (APIs) about image processing. In this paper, we design a recognition assisted surveillance system based on OSGi and OpenCV platforms. The system features dynamic monitoring by a camera carried by a robot and a Java 2 Micro-Edition (J2ME) viewer on a mobile phone. With the assistance of image recognition techniques, the captured frames are adaptively reproduced for handheld phones in a limited bandwidth environment. The proposed adaptive pause time control mechanism can efficiently improve the synchronization relationship between captured and viewed frames across heterogeneous networks. The evaluation results show that the proposed scheme can save power for the moveable camera and have a shorter time delay between the captured and viewed frames.     

Designing an integrated driver assistance system using image sensors

Road accidents cause a great loss to human lives and assets. Most of the accidents occur due to human errors, such as bad awareness, distraction, drowsiness, low training, and fatigue. Advanced driver assistance system (ADAS) can reduce the human errors by keeping an eye on the driving environment and warning a driver to the upcoming danger. However, these systems come only with modern luxury cars because of their high cost and complexity due to several sensors employed. Therefore, camera-based ADAS are becoming an option due to their lower cost, higher availability, numerous applications and ability to combine with other systems. Targeting at designing a camera-based ADAS, we have conducted an ethnographic study of drivers to know what information about the driving environment would be useful in preventing accidents. It turned out that information on speed, distance, relative position, direction, and size and type of the nearby objects would be useful and enough for implementing most of the ADAS functions. Several camera-based techniques are available for capturing the required information. We propose a novel design of an integrated camera-based ADAS that puts technologies??such as five ordinary CMOS image sensors, a digital image processor, and a thin display??into a smart system to offer a dozen advanced driver assistance functions. A basic prototype is also implemented using MATLAB. Our design and the prototype testify that all the required technologies are now available for implementing a full-fledged camera-based ADAS.     

基于OpenCV的管道巡检水下机器人的设计与实现

本设计是基于OpenCV的管道巡检水下机器人的设计与实现,水下机器人通过所配备的OpenCV摄像头对水下管道进行探测,主要应用于对水下或海底管道的路线进行循迹,并且能够对外壁破损情况的检查以及清理附着的杂物。该设计的实现是应用OpenCV摄像头对所需巡检的管道进行探测,并将所探测到的图像反馈到STM32控制器中识别出管道的循迹路线,并通过驱动模块中的电机设计实现平稳的循迹以及转弯功能。最终通过水下机器人的循迹以及图像处理功能的设计将探测到管道外部轮廓的图像信息进行处理,突出外壁所存在的问题以及附着的杂物。本设计将水下机器人驱动模式与图像处理功能进行很好的结合与应用,从而实现对管道循迹的过程中管道外壁的收检测。     

基于卷积神经网络的驾驶辅助系统设计

为了减轻驾驶员在行驶过程中的操作负担,进而降低误差判断事件的出现几率,设计一种基于卷积神经网络的驾驶辅助系统。在执行良好的汽车导航架构中,限定Learning Navigation模块与Learning Controller模块的连接位置,再根据辅助驾驶传感器对于行驶画面的采集情况,对车辆巡航能力进行定向控制,抑制监测仪表中辅助波的过渡振动,完成驾驶辅助系统的需求与设计分析。在此基础上,确定辅助激活函数、约束仪表中的行车图像,建立标准化的卷积神经网络。按照驾驶辅助数据的学习结果,对其进行传输处理,进而连接驾驶辅助系统的Job请求,实现系统的顺利运行。利用卷积神经网络平台设计实车实验结果表明,应用驾驶辅助系统后,车辆监测仪表中辅助波振动幅度的最小值处于36-61Hz之间,平均波长偏移量明显减小,驾驶员的行驶操作负担得到有效缓解。     

低照度条件下超分辨率人脸图像采集系统设计与实现

在计算机技术高速发展的时代,多平台计算机视觉库随之产生。OpenCV作为一种开源代码的计算机视觉库,以可兼容多平台、接口广泛的特点被广泛运用各个领域。在低照度条件下,会出现光照环境差异过大或光线不足等情况,导致传统图像采集系统不能采集高质量的人脸图像,局限性较差。提出基于OpenCV在C 环境配置下运用三维人脸识别技术算法,设计一套低照度条件下超分辨率人脸图像采集系统。实验证明,该设计方案具有实时(对焦速度快)、快速(单张采集0.05秒)、准确(面部识别率99.3%)等特点,能够充分满足低照度条件下超分辨率人脸图像采集的需求。     

翻转机构驱动力的优化及Adams仿真验证

在某航天器装配平台中,需要设计一个翻转机构实现部件与总体结构的对接工作,其中,外舱板翻转机构的驱动系统设计是重要内容之一.从外舱板翻转系统中外舱板翻转机构驱动力的设计出发,通过对特定负载的分析和理论分析,得到驱动力的大小以及驱动系统安装的最优位置.采用Adams的优化分析功能,以驱动系统安装位置为设计变量,以外舱板翻转机构在翻转过程中所需的最大驱动力中的最小值为目标函数,进行驱动系统的最优化仿真分析.应用Adams对优化设计结果进行仿真分析验证,得到最优结果.     

Shared steering control: How strong and how prompt should the intervention be for a better driving experience?

The lane keeping assistance system, a representative advanced driver assistance system, comprises a shared control that cooperates with the driver to achieve a common goal. The steering experience of the driver may vary significantly depending on the auto-steering control strategy of the system. In this study, we examined the driving experience with various steering control strategies. Nine control strategies (three torque amounts × three deviations in starting control) were established as prototypes. Eighteen drivers participated in the evaluation of each strategy in a highway environment on a driving simulator. A two-way repeated measure ANOVA was used to assess the effects of the system. Both the objective measures (standard deviation of lane position, steering reversal rate, and root mean square of lateral speed) and subjective measures (pleasure and arousal of emotion, trust, disturbance, and satisfaction) were evaluated and analyzed. The results showed that a torque amount of 3 Nm evoked feelings of high disturbance and negative emotional responses. A deviation in starting control (DEV) of 0.80 m yielded unstable lane keeping performances and evoked negative effects on pleasure, trust, and satisfaction. A regression model for the driver satisfaction recommended a torque of 2.32 Nm and a DEV of 0.27 m as the optimal design parameters. This proposed strategy is expected to improve the experience design of lateral semi-autonomous vehicles.     

A verified Common Lisp implementation of Buchberger’s algorithm in ACL2

In this article, we present the formal verification of a Common Lisp implementation of Buchberger’s algorithm for computing Gröbner bases of polynomial ideals. This work is carried out in ACL2, a system which provides an integrated environment where programming (in a pure functional subset of Common Lisp) and formal verification of programs, with the assistance of a theorem prover, are possible. Our implementation is written in a real programming language and it is directly executable within the ACL2 system or any compliant Common Lisp system. We provide here snippets of real verified code, discuss the formalization details in depth, and present quantitative data about the proof effort.     

An Intelligent and Integrated Driver Assistance System for Increased Safety and Convenience Based on All-around Sensing

Advanced driver assistance systems (ADAS) support the driver’s decision making to increase safety and comfort by providing an ergonomic display of the driving environment as well as issuing the warning signals or even exerting active control in case of dangerous conditions. Most previous research and products intend to offer only a single warning service, such as lane departure warning, collision warning, lane change assistance, etc. Although each component of these functions elevates the driving safety and convenience to a certain degree, a new type of ADAS will be developed to integrate all of the important functions with an efficient human–machine interface (HMI) framework for various driving conditions. We present an all-around sensing system based on an integrated ADAS that senses all directions using 2 cameras and 8 sonars, recognizes the driving environment via lane and vehicle detection, and construct a novel bird’s-eye view HMI of the environment for easy comprehension that even gives a proper warning signal in case of imminent danger. It was tested on our experimental vehicle with a good demonstration its working. Further, it has a good potential for commercial use by virtue of the low cost of the sensors used.     

基于OpenCV的运动目标定位跟踪系统软件设计

目标定位跟踪技术一直是计算机视觉领域的一个研究热点。为了更好地改进目标跟踪算法的跟踪特性,设计一种基于OpenCV的目标定位跟踪软件系统。通过对Camshifl跟踪算法的研究和改进,以实际的跟踪环境为背景,利用VS2008软件平台和OpenCV库函数,设计一种可以控制云台并实时定位跟踪运动目标的软件系统。实验证明,本系统定位跟踪精确,为运动目标的定位跟踪的研究和应用提供了方便,具有一定的实际意义。     

基于参考点配准法的全景泊车系统研究

为了提高驾驶行车的安全性,研究并改进了一种全景泊车系统,由鱼眼摄像头,图像处理单元和显示器三大部分组成.鱼眼摄像头采集汽车四周的超广角图像;在OpenCV环境下通过鱼眼摄像头标定,桶型矫正和图像的透视变换技术形成汽车前后左右的4幅平面俯视图;提出了一种参考点配准的方法,可以根据平面俯视图得到汽车的360°全景图像,并在显示器上显示.与特征点匹配法相比,参考点配准方法受环境的影响很小且对硬件性能要求较低,因此,更适用于嵌入式系统实现.     

缩微智能车室内定位系统研究

为在缩微交通环境下模拟实现现实交通中GPS对车辆的定位以及辅助驾驶功能,文中利用RFID以及网络通信技术建立了缩微智能车的室内定位系统,并在此基础上运用卡尔曼滤波算法对缩微智能车进行机动目标跟踪,实现了缩微车在缩微交通环境下的实时定位显示。大量实验结果表明,该室内定位系统能够可靠显示智能车的当前位置,定位误差小于0．07m,满足缩微交通环境下缩微车的定位需求。     

实时系统的形式化验证

实时系统的设计对系统设计人员而言是一个巨大挑战。在缺乏严格的验证环境时 ,要避免设计错误是很困难的。本文将一种带时戳的时序逻辑及用于描述具体实时系统的时间变迁系统编码到 HOL定理证明器中 ,并实现了一个基本的规则策略库 ,从而实现了一个简单的交互式辅助验证环境L RP。实例 Fisher算法的互斥性在 IRP中得到了验证。     

一种图像处理和计算机视觉的开发工具

该文探讨了可用于图像处理与计算机视觉编程的强大类库OpenCV，该类库使用起来极为方便，利用OpenCV中的数字图处理和计算机视觉的函数处理相关问题变得很简单。该文首先介绍OpenCV的强大功能以及研究的意义，然后介绍了OpenCV新版本的一些特点，并且讨论在VC＋＋环境下的软件设置问题，最后给出了典型的图像处理的实例。随着计算机视觉和数字图像处理技术不断深入各个领域，OpenCV为VC＋＋编程处理数字图像提供了极大的方便，具有广阔的应用前景，该文对于图像处理与计算机视觉方面的应用设计以及研究开发都将具有重要参考价值。     

基于RTEMS的软件容错系统设计研究

运行于空间环境中的计算机系统，受空间辐射的影响，可能会发生各种各样的错误。为了提高系统可靠性，需要对其进行容错。在本文中，首先分析RTEMS本身的特点和空间环境的特点；然后在此基础上提出基于RTEMS的软件容错系统的分层设计思想，并给出具体的层次设计过程；最后结合模拟实验验证给出容错状态下系统的性能分析。实验结
果表明，对容错系统的分层设计是必要的。     

基于GPU的车载全景视觉系统

针对大、中型车辆在行驶、泊车过程中存在的视觉盲区以及多路视频全景拼接难以实时等工程化问题,本文设计出一种基于FPGA和GPU平台的多路相机安全驾驶辅助系统.FPGA实现图像数据采集、参数传递等预处理,在GPU上实现拼接算法的并行加速.在算法上优化实现多相机自动化标定,生成融合参数表以获取精确的融合配准关系.实验结果表明,该系统可适应多路鱼眼相机的实时拼接,在TX2上稳定的拼接速度可达到33 fps.     

利用 OpenCV 实现在 Android系统上的图像边缘检测

论文提出了利用OpenCV实现在Android系统上图像边缘检测的方法。介绍了Android NDK开发环境的搭建,利用JNI调用OpenCV中的Canny算子,采用Android NDK生成动态链接库。实验结果表明在Android系统上图像边缘检测功能和性能良好。     

DoMAIns: Domain-based modeling for Ambient Intelligence

Ambient Intelligence and Smart Home Automation systems are currently emerging as feasible and ready to exploit solutions to support more intelligent features inside future and current homes. Thanks to increased availability of off-the-shelf components and to relatively easy to implement solutions we are experiencing a steady evolution of households, causing an ever-increasing users’ awareness of the capabilities of such innovative environments. To foster effective adoption of Smart Home Automation technologies in our home environments, traditional architectural and plant design must be complemented by sound design methodologies and tools, supporting the whole environment design cycle, including for example modeling, simulation and emulation, as well as, when feasible, formal model-checking and verification. Several research efforts have already addressed the design of expressive modeling tools, mostly based on Semantic Web technologies, as well as of suitable platforms for adding interoperation and rule-based intelligence to home environments. This paper proposes a new modeling methodology designed to fit the different phases of Intelligent Environments design, with a particular focus on validation and verification of the whole system. Carefully designed separation of modeled entities permits to exploit the DoMAIns framework during all phases of the environment design, from early abstract conception to the final in-field deployment. The DoMAIns design methodology is applied to a sample use case that involves comprehensive modeling and simulation of a Bank Security Booth, including the environment, the control algorithms, the automation devices and the user. Results show that the approach is feasible and that can easily handle different types of environment modeling, required in the different design phases, and for each of them it may support simulation, emulation, or other verification techniques.     

Examining System Administrators’ Verification Information to Enhance IT Training Design

System administrators are faced with a broad range of responsibilities for highly complex environments. They need the ability to verify the work they have done. That verification comes from accurate and available information that we refer to as information credibility. This exploratory research aims to address the relationship between task complexity, task risk, and verification information seeking in graphical user interface (GUI) tools used by system administrators. This research also illustrates the use of verification information as a starting point for a content-centered training design. Potential antecedents of verification information are identified and a model is proposed that addresses how aspects of the task and environment affect the need for verification. Findings suggest that task complexity is a significant indicator of the need for verification information. In addition, findings also suggest that this verification information provides a logical foundation for improving knowledge acquisition of the system administration team.