Real-Time Disk Scheduling for Multimedia Applications with Deadline-Modification-Scan Scheme

Real-timedisk scheduling (RTDS) is important for time-critical multimediaapplications. In conventional approaches of real-time disk scheduling,such as SCAN-EDF, the seek-optimizing SCAN scheme is appliedto reduce the disk service time of the real-time EDF schedule.Since only tasks with the same deadline are seek-optimized, theobtained improvement of SCAN-EDF is limited. In this paper, basedon the maximum-scannable-group (MSG) concept, a deadline-modification-scan (DMS or DM-SCAN) algorithm is proposed.Our algorithm uses MSG to automatically decide the suitable taskgroups for seek-optimizing. Based on the MSG concept, we repeatedlyapply DMS to further increase disk throughput to support moretasks. We have implemented the proposed algorithm on UnixWare2.01. The appropriate problem behaviors and parameter valuesto yield good solutions are investigated. Experiments show thatDMS is better than conventional approaches in both the obtaineddisk throughput and the supported tasks. Moreover, our proposedapproach can schedule task sets that are not schedulable by EDFand SCAN-EDF.     

Real-Time Multimedia Applications in a Web-Based Robotic Telecare System

In this paper, real-time multimedia applications are integrated into a Web-based robotic telecare system. Using Java media framework (JMF), a video/audio conference system is developed to improve the interaction among caregivers, and switches remote robot manipulating privilege with the help of a centralized management server. This video/audio conference system is established on a similar profile of H.323 multipoint control unit (MCU). Access control is utilized to guarantee confidentiality of telecare users. Different telecare services are provided to meet requirements of various users. The whole telecare system is verified in a campus network.     

The Pipe-group Architecture for Real-Time Active Vision

This paper describes a vision processing architecture which, together with a robotic camera platform and control subsystem, comprise a multi-processor real-time system for research into active vision. Experimentation in time-critical aspects of the active vision paradigm over several years has demonstrated the flexibility of ‘pipe-groups’ used as the building block in the system. The design of the entire system is overviewed, and the design of pipe-groups and their software is described in more detail. An active method of determining the exact processing latency, required for stable control, is explained. The paper illustrates the use of the pipe-group processing architecture and its real-time processing capabilities, with an example of active stereo tracking and structure recovery using the method of affine tranfer.     

嵌入式系统上的实时语音识别算法

介绍了一种嵌入式系统上的孤立词语音识别算法.该算法基于连续隐含马尔可夫模型,根据嵌入式系统的特点,简化了经典的连续隐含马尔可夫模型,在主流个人数字助理（PDA）上实现了中等规模语音识别的实时处理,采用最大后验概率（MAP）自适应方法解决训练数据采集信道和PDA信道的不匹配问题.在系统的后端处理中,提出了一种基于置信测度的拒识方法改善系统的稳健性,最终使610个孤立词的识别任务,系统的等错误率小于5%;对集内发音拒识率为5%时,集内发音识别率达到95%.     

基于视频、音频和文本的视频分段

在分析应用视频数据的过程中,视频分段是分析,组织,应用视频数据的基础。由于视频数据的多样性,传统的分段方法不能给出令人满意的结果,一般需要通过人机交互来进行。文中将较为成熟的文本分析、语音处理、图像处理三种技术进行综合,互为补充,对视频流进行分割。文本分析的对象是语音转换成的文本、标题、注释等。语音处理包括语音识别和语音信号分析。语音识别将视频中的自然语言转换为文字。语音信号分析对视频材料中的语音成分进行基础分析。图像处理主要用来处理视频中的图像部分。文章阐述了视频流的分段层次,文本分析,语音处理算法以及镜头突变,镜头渐变识别算法的思想。     

Using Real-Time Stereo Vision for Mobile Robot Navigation

This paper describes a working vision-based mobile robot that navigates and autonomously explores its environment while building occupancy grid maps of the environment. We present a method for reducing stereo vision disparity images to two-dimensional map information. Stereo vision has several attributes that set it apart from other sensors more commonly used for occupancy grid mapping. We discuss these attributes, the errors that some of them create, and how to overcome them. We reduce errors by segmenting disparity images based on continuous disparity surfaces to reject spikes caused by stereo mismatches. Stereo vision processing and map updates are done at 5 Hz and the robot moves at speeds of 300 cm/s.     

Real-Time Global Illumination for VR Applications

Real-time global illumination in VR systems enhances scene realism by incorporating soft shadows, reflections of objects in the scene, and color bleeding. The Virtual Light Field (VLF) method enables real-time global illumination rendering in VR. The VLF has been integrated with the Extreme VR system for real-time GPU-based rendering in a Cave Automatic Virtual Environment.     

一种针对多媒体扩展指令集和实际多媒体程序的自动向量化方法

自动向量化编译是利用处理器的多媒体扩展指令集提升多媒体程序性能的理想工具．但目前的研究不能有效加速实际程序．其主要原因是：普通算术操作的向量化不一定有性能提升;而多媒体典型操作因为其在源代码中表现形式多样而不能充分向量化．为了解决这一问题,文章对经典向量化算法进行改进以灵活统一地向量化这两类操作．主要的改进是增加了两个步骤：统一操作的不同表现形式和识别有价值的向量化操作．改进后的算法可以充分利用指令集生成高效代码,从而对实际多媒体程序拥有良好效果．此外,该算法可扩展性也很强．     

A Model for Updating Real-Time Applications

Updating application software is a common occurrence for modern computing systems. Software updates stem from the need to correct coding errors or to enhance the functionality of an application. Updating an application typically requires taking the current application offline and restarting a new application. This method of updating an application is perfectly acceptable for many general purpose-computing environments. However, in real-time environments that require high availability and have stringent timing constraints, taking a process offline for updates may be unacceptable or pose unnecessary risks. Some examples of these environments include telecommunications, air traffic control, railway control and medical patient monitoring. We present a new method to dynamically update a real-time application without having to take it offline. Our new method, which we call dynamic update for real-time systems, can be used to update real-time applications using rate-monotonic scheduling, while preserving the original deadline guarantees.     

An Open Environment for Real-Time Applications

This paper describes an open system architecture that allows independently developed hard real-time applications to run together and supports their reconfiguration at run-time. In the open system, each real-time application is executed by a server. At the lower level, the OS scheduler schedules all the servers on the EDF basis. At the upper level, the server scheduler of each server schedules the ready jobs of the application executed by the server according to the algorithm chosen for the application. The paper describes the two-level CPU scheduling scheme used by the open system and the design and implementation of a uniprocessor open system within the framework of the Windows NT operating system. The implementation consists of three key components: the two-level hierarchical kernel scheduler, common system service providers, and real-time application programming interface.     

A Distributed System for Real-Time Applications

This versatile network uses multiple microprocessors and a split-bus architecture to collect and process real-time data from a variety of sources and then transfer it to different destinations.     

A Testbed for Mobile Multimedia Applications

This article describes the architecture, design and implementation of a mobile network emulator able to enhance development of mobile multimedia applications. In particular, it generates test situations which otherwise could be encountered only by moving around the equipment to be tested.     

光互连技术及其在计算机视觉中的应用

可重构的光互连技术具有针对不同的应用提供相应的网络结构的能力。在大规模并行系统中具有很大的优势,可以作为电互连的替代技术应用于计算机视觉这一类计算机和通信速度很高的系统中。     

Color Image Segmentation for Multimedia Applications

Image segmentation is crucial for multimedia applications. Multimedia databases utilize segmentation for the storage and indexing of images and video. Image segmentation is used for object tracking in the new MPEG-7 video compression standard. It is also used in video conferencing for compression and coding purposes. These are only some of the multimedia applications in image segmentation. It is usually the first task of any image analysis process, and thus, subsequent tasks rely heavily on the quality of segmentation. The proposed method of color image segmentation is very effective in segmenting a multimedia-type image into regions. Pixels are first classified as either chromatic or achromatic depending on their HSI color values. Next, a seed determination algorithm finds seed pixels that are in the center of regions. These seed pixels are used in the region growing step to grow regions by comparing these seed pixels to neighboring pixels using the cylindrical distance metric. Merging regions that are similar in color is a final means used for segmenting the image into even smaller regions.     

Non-Preemptive Real-Time Scheduling of Multimedia Tasks

Motivated by the special characteristics of multimedia tasks, we consider non-preemptive scheduling of tasks where there exists no (or very limited) information concerning the tasks before they are released. We present impossibility results and analyze algorithms for non-preemptive scheduling in single processor and multiprocessor systems. To evaluate our algorithm we assume that system obtains a value that is proportional to the processing time of the task whenever a task is completed by its deadline. Competitive analysis is used, where the goal is to keep the total value obtained by an on-line algorithm bounded by a function of the total value obtained by an off-line algorithm. In particular, one set of our results considers the competitive ratio of scheduling algorithm when the length of the tasks is not greater than C_max (and not smaller than C_min ). We show that the performance of a scheduling algorithm is improved dramatically when the release time of the tasks is O(C_max) prior to their deadline; achieving a competitive ratio that is close to one.     

Differentiated Real-Time Data Services for E-Commerce Applications

The demand for real-time e-commerce data services has been increasing recently. In many e-commerce applications, it is essential to process user requests within their deadlines, i.e., before the market status changes, using fresh data reflecting the current market status. However, current data services are poor at processing user requests in a timely manner using fresh data. To address this problem, we present a differentiated real-time data service framework for e-commerce applications. User requests are classified into several service classes according to their importance, and they receive differentiated real-time performance guarantees in terms of deadline miss ratio. At the same time, a certain data freshness is guaranteed for all transactions that commit within their deadlines. A feedback-based approach is applied to differentiate the deadline miss ratio among service classes. Admission control and adaptable update schemes are applied to manage potential overload. A simulation study, which reflects the e-commerce data semantics, shows that our approach can achieve a significant performance improvement compared to baseline approaches. Our approach can support the specified per-class deadline miss ratios maintaining the required data freshness even in the presence of unpredictable workloads and data access patterns, whereas baseline approaches fail.     

多媒体视觉VR重构算法的改进与仿真

传统图像重构噪声偏大,图像离群点较多,导致重构效果与实际图像偏差大,研究提出多媒体视觉VR重构算法。利用双目偏移定位重新校正VR图像,对重构后的VR图像进行去噪处理,完成基于高通滤波矩阵加强的重构图像纹路细节特征,采用三维重构相关性检索方法,获得合群点与离群点,确定其最大最小深度范围,对错误识别和峰值进行重新定位,完成算法设计。为验证研究的有效性,进行实验验证。通过实验结果数据分析可证明,研究算法大大提升了重构图像的真实度,图像角点检测准确,重构后图像噪声更小。     

Acoustically Driven Programmable Microfluidics for Biological and Chemical Applications

In summary, we described a novel and unconventional technique to manipulate smallest amounts of liquid on a chip. Using SAW on a piezoelectric substrate, we are able to actuate individual droplets along predetermined trajectories, or induce acoustically driven internal streaming in the fluid. This internal acoustic streaming can efficiently be used to agitate, mix, and stir very small liquid volumes, where the low Reynold's number usually only allows for diffusive mixing. We described several applications of the SAW driven microfluidics, including a nanomixer for microarray applications, a contactless mixer for MTPs, a programmable microfluidic chip for droplet-based assays, and finally a chip performing high resolution microliter PCR. The technique is equally well suited to actuate or agitate small amounts of liquids either in closed volumes or in an open, droplet-based geometry. Each of the approaches has its clear advantages, but also disadvantages. Droplet-based fluidics is certainly well suited to handle smallest amounts of fluids without the risk of cross contamination etc. High temperature processes, however, require additional means like covering the droplet with an oil film. SAW pumping, mixing, and stirring on closed volumes is advantageous over many other pumping schemes as the pumps are easily incorporated into most of the existing microfluidic device or lab-on-a-chip. The combination of the SAW actuated droplet-based fluid handling and SAW driven fluidics in closed volumes opens a wide field of many different applications, a few of which I had the pleasure to present in this article. Many more applications, and many more visualizations of the technology described above can be looked up on Advalytix' website http://www.Advalytix.de.