Real-Time Implementation of a New Contour Tracking Procedure in a Multi-Processor DSP System.

Size: px
Start display at page:

Download "Real-Time Implementation of a New Contour Tracking Procedure in a Multi-Processor DSP System."

Transcription

1 Real-Time Implementation of a New Contour Tracking Procedure in a Multi-Processor DSP System. V. GEMIGNANI (1), M. DEMI (1)(2), M. PATERNI (1), A. BENASSI (1) (1) CNR - Institute of Clinical Physiology, v. Savi 8, Pisa (2) Esaote SpA, Florence ITALY Abstract: - In this paper the implementation of a contour tracking procedure is presented. If a rough contour of the desired structure is available on the first frame of a sequence, the contours on the subsequent frames are automatically outlined in real time. Furthermore, indices of clinical interest are calculated from the computed contour and they are graphically displayed in real time. The software runs in an integrated software/hardware environment based on the interaction of a PC and a multi-processor DSP (TMS320C80). The paper describes the solutions adopted in order to obtain an efficient software implementation. The performance of the system was investigated and a comparison with a PC implementation is reported. The system was tested by monitoring the area changes of the thoracic aorta in sequences of echographic images. Key-Words: real-time, contour tracking, DSP, medical imaging, multi-processor. 1 Introduction In recent years, there has been increasing interest in the design and implementation of systems for real time contour tracking [1][2]. In particular, in medical imaging, a real time quantitative analysis of sequences of images is of considerable clinical interest [3][4]. Our task is to compute the contour of the structure under examination automatically, to estimate the indices of interest and to display the results in real-time. In a previous paper [5], a new automatic contour tracking procedure was presented and tested over sequences of cardiovascular images. Moreover, in [6] the TMS320C80 multi-processor DSP [7][8] proved to be a suitable device for obtaining a real time execution of the new contour tracking algorithm. In this paper, a Windows NT based PC equipped with a TMS320C80 DSP board [9] is adopted to realize the real time image processing system. The PC provides a suitable environment for the development of a flexible and powerful user interface. The DSP board provides the processing capability, the image capture and the image display resources necessary to obtain real time performances. The aim of this paper is to illustrate the solutions we adopted in the implementation of the procedure. The most onerous part of the work was the implementation of the algorithms on the DSP. In fact, the new high performance DSPs adopt special architectures which make it difficult for a compiler to produce an efficient code. Therefore, these devices are typically programmed in assembly language and efficient programming is actually a difficult task. Several library routines are available to simplify the implementation work. These routines can be used as building blocks to assemble more complex applications in a quicker manner [10]. Furthermore, in [11] a multi-layered library was developed to decrease the overhead of the resultant application. However, in this paper the algorithm was completely implemented without the use of libraries. This solution allowed a higher optimization of the code. Sequences of images of the descending thoracic aorta recorded by transesophageal echocardiography were used to test the system. The cross-sectional area of the aorta was computed in real time and plotted against time on the user interface. 2 Algorithm Overview

2 A new contour tracking procedure based on the FOAM (first order absolute moment) mass center operator was presented in [5][6]. In this paragraph, an overview of the procedure is reported. Let f(n, be the gray level map of an image, then the generalized first order absolute moment is computed as follows: e( n, = f1( n, f ( n m g2( Θ2 where f1( n, = f ( n m g1( Θ1 where Z represents the integer numbers and g i ( are discrete Gaussian functions with standard deviation σ i defined as: if ( Θ i ; g i (=0 elsewhere. Let Γ be a discrete vector with components l; then the FOAM mass center vector is defined as follows: 1 b( n, = e( n, Θi = { Z : k + l 3σ } gi( = Θ Θ i 2 if e(n, 0; b(n,=0 elsewhere. 1 f ( n, f ( n m à g 2( ) 1 l Let us consider a step straight discontinuity and let us compute the vector b at a point p at a distance d from the discontinuity. If d<3σ 2 and if the right configuration (σ 1,σ 2 ) of the filter is chosen, the vector b provides a mass center p which is closer to the discontinuity than p independently of the distance d [5]. Therefore, if an approximate starting contour C S is traced manually by an operator on the first frame of an image sequence, the true contour C 1 can be estimated by iteratively computing the FOAM mass center at the points of C S. Three iterations were enough to locate the contour in the synthetic and clinical images used to test the procedure. Once C 1 is computed, it is used as the i ( 2 2 k + l ) 2 2σ i 2 2 ( ) e k + l 2 2 σ i e approximate starting contour to compute C 2 in the second frame and so on. Therefore, if a starting contour is available on the first image of the sequence, the procedure automatically outlines the contours on the subsequent images. The computational cost of the procedure is primarily due to the iterative FOAM mass center computation. However, the algorithm and the number of steps are the same for all the points of the contour and the FOAM is computed separately at every single point. Therefore, the same algorithm is repeated several times with different input data. This allows an efficient parallel implementation in a SPMD (single program multiple data) scheme. 3 Implementation The system is based on the interaction between a single-cpu PC and a multi-cpu DSP board. In all, 6 CPUs are present: I. One general purpose processor (Pentium MMX). II. One RISC processor, the Master Processor (MP). III. Four DSP processors, the Parallel Processors (PPs). Due to the different architectures of the above mentioned processors, it is necessary to use three different implementation levels and three different software development tools (figure 1). Personal Computer DSP Board MP Contour data PP0 PP1 PP2 PP3 Host level MP level PP level Figure 1: Six CPUs and three levels of software implementation are present in the system. 3.1 Host level The Pentium-based PC can be efficiently programmed by using a high-level language. In this paper, the Microsoft Visual C language was adopted. The main function of the software running on the PC is to act as a user interface. It allows the operator to set-up the parameters of the filters, and to start and stop both the processing and the display

3 activities. Furthermore, it allows the operator to trace the starting contour on the first frame of the sequence by using the mouse. Once the processing activity has started, the PC continuously receives the data of the contours which are calculated by the DSP board. A new contour is received from the DSP board every 40ms and it is displayed in realtime in a special window. The computation of indices such as area and perimeter was implemented at the host level. This solution was chosen because the computational cost of these algorithms is not high, so they can be easily executed in real time by using a high level system. Moreover, even if the contour tracking procedure is always the same, the computation of specific indices is necessary for every application. Therefore, the use of the same DSP software with different PC programs proves to be a more flexible solution. 3.2 MP level The MP coordinates the on board processing activities and the communication with the PC. The software of the RISC MP was written in C and it runs in a multitasking environment. Three tasks were implemented (figure 2): Initialization Wait Initialization Task Wait for a command from the PC Process the command T_REG T_TRANSF Host Communication Task Image transfer from input to output buffer Regularization Wait until a new frame is captured PP start processing command Wait for PP end-ofprocessing Video Processing Task Figure 2: Three tasks were implemented at the MP level implementation. Initialization task: The task only operates when the board is started up. The program initializes the on board resources and the constants which are used by the PPs. Host communication task: This task manages the communication between the DSP board and the PC. The MP is configured as a server so it continuously waits for a command from the host. Typical commands regard the configuration of the filter, the start and stop of data acquisition and processing, the management of the mouse pointer, etc. Video task: The video task manages the video processing activity which needs to be synchronized with the data flow. For this purpose, before starting a new processing iteration, the MP waits until a new frame is captured in memory. Afterwards, all the processing activities need to be completed in 40ms to obtain a real time execution. The contour tracking procedure requires two distinct algorithms. The first algorithm computes the FOAM mass center and it is implemented at the PP level. In this case, the role of the MP is limited to coordinating the PPs which work in parallel in a SPMD scheme. The second one is a regularization algorithm which operates on the points of the contour. The computational cost of this algorithm is not onerous because of the small amount of data. However, the implementation is demanding because of the complexity of the algorithm. Moreover, more than one algorithms must be implemented to provide a configurable regularization. Therefore, the best solution in terms of speed, simplicity and flexibility proved to be the implementation of the algorithms on the MP using a high level language. Once the new contour is computed, it is superimposed on the image data and the obtained image is transferred to the output buffer. To conclude, three operations are executed sequentially: the iterative computation of the FOAM mass center, the regularization process and the image data transfer. If we define the time necessary for the above mentioned operations T_FOAM, T_REG and T_TRANSF, respectively, the constraint in order to obtain real time processing at 25fps is: T_FOAM + T_REG + T_TRANSF < 40ms These values vary depending on the data size. However, the image size, the filter configuration and the number of points of the contours are constant and consequently, the contour tracking procedure requires a constant number of operations at every image of the sequence. Therefore, since the processing time is about the same for every image, the real time working capability can be easily verified by using the software development tools. 3.3 PP level Since the computation of the FOAM mass center is the most onerous part of the procedure, it was implemented at the PP level. This implementation

4 required a lot of work because its optimization greatly influences the performance of the whole procedure. The use of the assembly language is mandatory at the PP level to achieve reasonable performances. Moreover, if the data are in the offchip memory, the DMA coprocessor must be used. In fact, direct external memory access is very costly on the PPs (10~15 cycles per access in the best case) and the data caching is not performed. Therefore, before starting the processing activity, the data need to be downloaded in the internal memory where the access time is only 1 cycle. The DMA coprocessor permits automatic multidimensional data transfer called PT (Packed Transfer). The algorithm can be divided into two parts: I. The convolution to obtain the value f 1 (n,. II. The summation of the absolute values of the differences. The input data of the convolution algorithm are the constant matrix g 1 ( and a region of interest ROI_1 of the image. Since g 1 ( is a constant, it was permanently stored in the local memory of every PP where it is accessible at the fastest speed. Instead, the data of ROI_1 are located in the external memory. Therefore, a PT is necessary in order to transfer the data to the internal memory before the data processing can be started. The inner loop of the procedure is two instructions long. Nevertheless, at every loop iteration two MACs (Multiply and Accumulate) are executed. Therefore, one multiplication and one addition are carried out per cycle. As regards the summation algorithm, the input data are again a constant matrix g 2 ( and a ROI_2 of the image. g 2 ( is permanently stored in the local memory whereas the ROI_2 is transferred from the external memory to the internal memory by using the DMA coprocessor. The inner loop of the procedure, which is 13 instructions long, processes four pixels per iteration. If the FOAM mass center is computed at two near points, the two ROI_1 and the two ROI_2 partially overlap. Therefore, when the ROI_i of the second point have to be transferred, some data are already present in the internal memory. We can either reload the whole ROI_i or introduce a more suitable mechanism which is able to re-use the previous data. The first approach was preferred because it simplifies and speeds up the subsequent processing phase. Moreover, when processing a single point, ROI_1 is always included in ROI_2. Nevertheless, two distinct PT are used to transfer the two ROI_i. This solution was adopted because every PT rearranges the organization of the data to speed up the execution of the relative processing algorithm. In particular, the management of the pointers and the loop controllers can be simplified by means of an optimized data arrangement. The adopted solution increases the data transfer time but greatly optimizes the implementation of the algorithm. The two functions g i ( have non-zero values only in the domains θ i so the two ROI_i should also be circular. Nevertheless, managing circular regions is a difficult task. Therefore, the ROI_i are square regions which include the θ i. The size of the ROI_i is n i xn i where n i =int(6 σ i )+1. In table 1 the size of the data and the number of operations of the algorithms are summarized. (a) Convolutio n (b) Summation Convolutio n Summation 2x(n 2 xn 2 )+n Input data Output Constant Variable data g 1 : [n 1 xn 1 ] ROI_1 : [n 1 xn 1 ] f 1 : [1x1] g 2 : [n 2 xn 2 ] ROI_2 : [n 2 xn 2 ] b : [2x1] MAC Subtr. Abs. Divis. n 1 xn n 2 xn 2 n 2 xn 2 2 Table 1: (a) The data and their size are summarized for both parts of the algorithm. (b) The number of MACs, subtractions, absolute values and divisions are reported as a function of the data size. 3.4 Arithmetic The processing speed of the TMS320C80 is primarily due to the performances of the four PPs. Each PP is an advanced 32-bit fixed-point DSP with special features to improve processing capability. In fact, they offer the possibility of splitting the data unit to process one 32-bit operation, two 16-bit operations or four 8-bit operations. The multiplier can also execute one 16-bit operation or two 8-bit operations. Therefore, the data representation

5 heavily influences the performance of the procedure. If a TMS320C80 is adopted, a fixed-point arithmetic must be mainly used. In fact, although the multiprocessor we used offers the remarkable floating point performance of up to 80 MFLOPS (at 40MHz), the device is optimized for fixed-point arithmetic. Moreover, because of its complexity, the use of this DSP is justified only if its resources are efficiently exploited. In the computation of the FOAM mass center, the input data are 8-bit integer values (256 gray level images are used). The masks g 1 ( and g 2 ( were also implemented as 8-bit integer values. Therefore, in the convolution algorithm, the multiplication of 8-bit data and the accumulation of 16-bit data were performed. Subsequently, the value f 1 (n,, which is computed as a 16 bit value, is rounded to an 8-bit value. Consequently, the input data in the summation algorithm are 8-bit wide and the splitting capability of the data unit and of the multiplier can be used again efficiently. Moreover, the SPMD scheme where the four PPs work in parallel under the management of the MP is exploited. To conclude, the computation of the FOAM mass center in 8-bit gray level images proved to be a task which efficiently matched the architecture of the DSP we adopted. 4 Results and Conclusions The performances of the system were tested with several filter configurations. Let us consider the case in which three iterations of the FOAM mass center are computed at 40 points of a contour and the configuration of n 1 =17pixels, n 2 =25pixels is adopted: the time required to compute the three iterations at all the points is T_FOAM=4.6ms. The measurement was obtained in run-time from the start-of-processing command of the PPs to the endof-processing notification. Therefore, the time required to compute the FOAM mass center using one PP was: 4.6ms / (40knots 3iterations) 4PPs = 153µs Let us now consider the PP assembler procedure only, without taking into consideration the ROI_i data transfer time. The time required to compute the FOAM mass center was 77.5µs. Therefore the computational time is roughly doubled in comparison to the time required to process the data once they are in the on-chip memory. Three regularization algorithms were implemented: the uniform spacing of the points of the contour, a mean filtering, and a median filtering. When a contour with 40 points was used, T_REG was less then 1.5ms for all the procedures tested. The T_TRANSF parameter depends on the size of the image data. In this paper, 8-bits gray level images of 512x512 pixels were used and the time T_TRANS necessary to transfer the data to the output buffer resulted in 4.4ms. In order to evaluate the advantages obtained thanks to the adoption of a DSP architecture, a comparison with a PC implementation is reported in figure 3. The results refer to the computation of the FOAM mass center. A PC Pentium II 350 MHz was used and the algorithm was implemented in C. In all the configurations tested, a considerable increase in the computational speed was obtained. For example, if n 1 =25pixels and n 2 =37pixels, the computational time is 1330µs for the PC and 54µs for the DSP. Therefore, the remarkable advantage obtained justifies the adoption of the DSP architecture and the onerous implementation. [us] Figure 3: Computational time of the FOAM mass center algorithm for both the PC and the DSP implementation. The configuration of the filter is σ 1 =(2/3)σ 2. PC n2 [pixels] In figures 4 and 5, the result of the monitoring of the aorta area changes in an echographic examination is reported. The images were captured with a resolution of 512x512 pixels and a frame rate of 25fps. The new system proved to be a versatile tool for the real time estimation of indices of medical interest. Our current efforts are directed towards increasing the robustness of the procedure as this seems to be the weakness of the system. In this paper, constraints on the shape of the contour have

6 not been introduced and a large flexibility in the contour tracking capability was obtained. However, the robustness of the system is lower when compared to other papers which introduced such constraints [3][4]. Figure 4: Sequences of images of the descending thoracic aorta recorded by transesophageal echocardiography were used to test the system. Figure 5: User interface. At the top of the screen, the menus and the toolbar are located. In the central windows, the contour (on the left) and the cross-sectional area changes of the aorta (on the right) are depicted in real time. At the bottom, minimum and maximum values of the area are reported. Echocardiographic Sequence. Proc. ICCV98, Bombay, India, 1998, pp [4] McEachen, J.C.II and Ducan,J.S. Shapebased tracking of left ventricular wall motion. IEEE Trans. Medical Imaging, 1997,16 (3), pp [5] M. Paterni, M. Demi, C. Morizzo, M. Kozakova, C. Palombo The First Order Absolute Moment in Contour Tracking. Proc. Computers in Cardiology, Lund, Sweden, 1997, pp [6] V. Gemignani, S. Provvedi, M. Demi, M. Paterni, A. Benassi A DSP-Based Real Time Contour Tracking System. Proc. of ICIAP99, Venice, Italy, 1999, pp [7] Texas Instruments TMS320c8x System-Level Synopsis, SPRU113, [8] K. Guttag, R.J. Gove, J.R. Van Aken, A single-chip multiprocessor for multimedia: The MVP, IEEE Computer Graph. Appl., 1997, 12 (6), pp [9] Texas Instruments TMS320c80 Software Development Board, SPRU178, [10] F. Lohier, L. Lacassagne A Generic Methodology for the Software Managing of Caches in Multi-processors DSP architectures. Proc. of ICASSP99, Phoenix, Arizona, 1999, vol. 4, pp [11] J.Kim, Y.Kim UWICL: A Multi-Layered Parallel Image Computing Library for Single- Chip Multiprocessor-based Time-Critical Systems. Real Time Imaging, 1996, 2, pp References: [1] M. Hernández, J. Cabrera, M. Castrillón, A. Domínguez, C. Guerra, D. Hernández, J. Isern DESEO: An Active Vision System for Detection, Tracking and Recognition. Proc. ICVS99, Las Palmas, Spain, 1999, pp [2] Z.Li, H.Wang Real time 3D Motion Tracking with Known Geometric Models. Real-Time Imaging, 1999, 5, pp [3] Gary Jacob, J. Alison Noble, Andrew Blake Robust Contour Tracking in

High-speed image processing algorithms using MMX hardware

High-speed image processing algorithms using MMX hardware High-speed image processing algorithms using MMX hardware J. W. V. Miller and J. Wood The University of Michigan-Dearborn ABSTRACT Low-cost PC-based machine vision systems have become more common due to

More information

FPGA. AT6000 FPGAs. Application Note AT6000 FPGAs. 3x3 Convolver with Run-Time Reconfigurable Vector Multiplier in Atmel AT6000 FPGAs.

FPGA. AT6000 FPGAs. Application Note AT6000 FPGAs. 3x3 Convolver with Run-Time Reconfigurable Vector Multiplier in Atmel AT6000 FPGAs. 3x3 Convolver with Run-Time Reconfigurable Vector Multiplier in Atmel AT6000 s Introduction Convolution is one of the basic and most common operations in both analog and digital domain signal processing.

More information

Speed Performance Improvement of Vehicle Blob Tracking System

Speed Performance Improvement of Vehicle Blob Tracking System Speed Performance Improvement of Vehicle Blob Tracking System Sung Chun Lee and Ram Nevatia University of Southern California, Los Angeles, CA 90089, USA sungchun@usc.edu, nevatia@usc.edu Abstract. A speed

More information

In: Proceedings of RECPAD 2002-12th Portuguese Conference on Pattern Recognition June 27th- 28th, 2002 Aveiro, Portugal

In: Proceedings of RECPAD 2002-12th Portuguese Conference on Pattern Recognition June 27th- 28th, 2002 Aveiro, Portugal Paper Title: Generic Framework for Video Analysis Authors: Luís Filipe Tavares INESC Porto lft@inescporto.pt Luís Teixeira INESC Porto, Universidade Católica Portuguesa lmt@inescporto.pt Luís Corte-Real

More information

Freescale Semiconductor, I

Freescale Semiconductor, I nc. Application Note 6/2002 8-Bit Software Development Kit By Jiri Ryba Introduction 8-Bit SDK Overview This application note describes the features and advantages of the 8-bit SDK (software development

More information

AN EVALUATION OF MODEL-BASED SOFTWARE SYNTHESIS FROM SIMULINK MODELS FOR EMBEDDED VIDEO APPLICATIONS

AN EVALUATION OF MODEL-BASED SOFTWARE SYNTHESIS FROM SIMULINK MODELS FOR EMBEDDED VIDEO APPLICATIONS International Journal of Software Engineering and Knowledge Engineering World Scientific Publishing Company AN EVALUATION OF MODEL-BASED SOFTWARE SYNTHESIS FROM SIMULINK MODELS FOR EMBEDDED VIDEO APPLICATIONS

More information

A Computer Vision System on a Chip: a case study from the automotive domain

A Computer Vision System on a Chip: a case study from the automotive domain A Computer Vision System on a Chip: a case study from the automotive domain Gideon P. Stein Elchanan Rushinek Gaby Hayun Amnon Shashua Mobileye Vision Technologies Ltd. Hebrew University Jerusalem, Israel

More information

Control 2004, University of Bath, UK, September 2004

Control 2004, University of Bath, UK, September 2004 Control, University of Bath, UK, September ID- IMPACT OF DEPENDENCY AND LOAD BALANCING IN MULTITHREADING REAL-TIME CONTROL ALGORITHMS M A Hossain and M O Tokhi Department of Computing, The University of

More information

Architectures and Platforms

Architectures and Platforms Hardware/Software Codesign Arch&Platf. - 1 Architectures and Platforms 1. Architecture Selection: The Basic Trade-Offs 2. General Purpose vs. Application-Specific Processors 3. Processor Specialisation

More information

Parallel Digital Signal Processing: An Emerging Market

Parallel Digital Signal Processing: An Emerging Market Parallel Digital Signal Processing: An Emerging Market Application Report Mitch Reifel and Daniel Chen Digital Signal Processing Products Semiconductor Group SPRA104 February 1994 Printed on Recycled Paper

More information

A System for Capturing High Resolution Images

A System for Capturing High Resolution Images A System for Capturing High Resolution Images G.Voyatzis, G.Angelopoulos, A.Bors and I.Pitas Department of Informatics University of Thessaloniki BOX 451, 54006 Thessaloniki GREECE e-mail: pitas@zeus.csd.auth.gr

More information

Developing applications under CODE COMPOSER STUDIO

Developing applications under CODE COMPOSER STUDIO Developing applications under CODE COMPOSER STUDIO 1. General Overview Code Composer Studio (CCS ) is a very efficient instrument for the fast development of applications that are written for the DSP families

More information

Optimizing matrix multiplication Amitabha Banerjee abanerjee@ucdavis.edu

Optimizing matrix multiplication Amitabha Banerjee abanerjee@ucdavis.edu Optimizing matrix multiplication Amitabha Banerjee abanerjee@ucdavis.edu Present compilers are incapable of fully harnessing the processor architecture complexity. There is a wide gap between the available

More information

Parallelism and Cloud Computing

Parallelism and Cloud Computing Parallelism and Cloud Computing Kai Shen Parallel Computing Parallel computing: Process sub tasks simultaneously so that work can be completed faster. For instances: divide the work of matrix multiplication

More information

FLOATING-POINT ARITHMETIC IN AMD PROCESSORS MICHAEL SCHULTE AMD RESEARCH JUNE 2015

FLOATING-POINT ARITHMETIC IN AMD PROCESSORS MICHAEL SCHULTE AMD RESEARCH JUNE 2015 FLOATING-POINT ARITHMETIC IN AMD PROCESSORS MICHAEL SCHULTE AMD RESEARCH JUNE 2015 AGENDA The Kaveri Accelerated Processing Unit (APU) The Graphics Core Next Architecture and its Floating-Point Arithmetic

More information

A General Framework for Tracking Objects in a Multi-Camera Environment

A General Framework for Tracking Objects in a Multi-Camera Environment A General Framework for Tracking Objects in a Multi-Camera Environment Karlene Nguyen, Gavin Yeung, Soheil Ghiasi, Majid Sarrafzadeh {karlene, gavin, soheil, majid}@cs.ucla.edu Abstract We present a framework

More information

A New, High-Performance, Low-Power, Floating-Point Embedded Processor for Scientific Computing and DSP Applications

A New, High-Performance, Low-Power, Floating-Point Embedded Processor for Scientific Computing and DSP Applications 1 A New, High-Performance, Low-Power, Floating-Point Embedded Processor for Scientific Computing and DSP Applications Simon McIntosh-Smith Director of Architecture 2 Multi-Threaded Array Processing Architecture

More information

Final Year Project Progress Report. Frequency-Domain Adaptive Filtering. Myles Friel. Supervisor: Dr.Edward Jones

Final Year Project Progress Report. Frequency-Domain Adaptive Filtering. Myles Friel. Supervisor: Dr.Edward Jones Final Year Project Progress Report Frequency-Domain Adaptive Filtering Myles Friel 01510401 Supervisor: Dr.Edward Jones Abstract The Final Year Project is an important part of the final year of the Electronic

More information

Teaching DSP through the Practical Case Study of an FSK Modem

Teaching DSP through the Practical Case Study of an FSK Modem Disclaimer: This document was part of the First European DSP Education and Research Conference. It may have been written by someone whose native language is not English. TI assumes no liability for the

More information

An Implementation Of Multiprocessor Linux

An Implementation Of Multiprocessor Linux An Implementation Of Multiprocessor Linux This document describes the implementation of a simple SMP Linux kernel extension and how to use this to develop SMP Linux kernels for architectures other than

More information

BDTI Solution Certification TM : Benchmarking H.264 Video Decoder Hardware/Software Solutions

BDTI Solution Certification TM : Benchmarking H.264 Video Decoder Hardware/Software Solutions Insight, Analysis, and Advice on Signal Processing Technology BDTI Solution Certification TM : Benchmarking H.264 Video Decoder Hardware/Software Solutions Steve Ammon Berkeley Design Technology, Inc.

More information

Next Generation GPU Architecture Code-named Fermi

Next Generation GPU Architecture Code-named Fermi Next Generation GPU Architecture Code-named Fermi The Soul of a Supercomputer in the Body of a GPU Why is NVIDIA at Super Computing? Graphics is a throughput problem paint every pixel within frame time

More information

Implementation of Canny Edge Detector of color images on CELL/B.E. Architecture.

Implementation of Canny Edge Detector of color images on CELL/B.E. Architecture. Implementation of Canny Edge Detector of color images on CELL/B.E. Architecture. Chirag Gupta,Sumod Mohan K cgupta@clemson.edu, sumodm@clemson.edu Abstract In this project we propose a method to improve

More information

Contributions to Gang Scheduling

Contributions to Gang Scheduling CHAPTER 7 Contributions to Gang Scheduling In this Chapter, we present two techniques to improve Gang Scheduling policies by adopting the ideas of this Thesis. The first one, Performance- Driven Gang Scheduling,

More information

A Survey of Video Processing with Field Programmable Gate Arrays (FGPA)

A Survey of Video Processing with Field Programmable Gate Arrays (FGPA) A Survey of Video Processing with Field Programmable Gate Arrays (FGPA) Heather Garnell Abstract This paper is a high-level, survey of recent developments in the area of video processing using reconfigurable

More information

Performance Analysis and Comparison of JM 15.1 and Intel IPP H.264 Encoder and Decoder

Performance Analysis and Comparison of JM 15.1 and Intel IPP H.264 Encoder and Decoder Performance Analysis and Comparison of 15.1 and H.264 Encoder and Decoder K.V.Suchethan Swaroop and K.R.Rao, IEEE Fellow Department of Electrical Engineering, University of Texas at Arlington Arlington,

More information

Shear :: Blocks (Video and Image Processing Blockset )

Shear :: Blocks (Video and Image Processing Blockset ) 1 of 6 15/12/2009 11:15 Shear Shift rows or columns of image by linearly varying offset Library Geometric Transformations Description The Shear block shifts the rows or columns of an image by a gradually

More information

Data Storage 3.1. Foundations of Computer Science Cengage Learning

Data Storage 3.1. Foundations of Computer Science Cengage Learning 3 Data Storage 3.1 Foundations of Computer Science Cengage Learning Objectives After studying this chapter, the student should be able to: List five different data types used in a computer. Describe how

More information

Automatic Traffic Estimation Using Image Processing

Automatic Traffic Estimation Using Image Processing Automatic Traffic Estimation Using Image Processing Pejman Niksaz Science &Research Branch, Azad University of Yazd, Iran Pezhman_1366@yahoo.com Abstract As we know the population of city and number of

More information

A Pattern-Based Approach to. Automated Application Performance Analysis

A Pattern-Based Approach to. Automated Application Performance Analysis A Pattern-Based Approach to Automated Application Performance Analysis Nikhil Bhatia, Shirley Moore, Felix Wolf, and Jack Dongarra Innovative Computing Laboratory University of Tennessee (bhatia, shirley,

More information

A Performance Study of Load Balancing Strategies for Approximate String Matching on an MPI Heterogeneous System Environment

A Performance Study of Load Balancing Strategies for Approximate String Matching on an MPI Heterogeneous System Environment A Performance Study of Load Balancing Strategies for Approximate String Matching on an MPI Heterogeneous System Environment Panagiotis D. Michailidis and Konstantinos G. Margaritis Parallel and Distributed

More information

Computer Architecture TDTS10

Computer Architecture TDTS10 why parallelism? Performance gain from increasing clock frequency is no longer an option. Outline Computer Architecture TDTS10 Superscalar Processors Very Long Instruction Word Processors Parallel computers

More information

Data Storage. Chapter 3. Objectives. 3-1 Data Types. Data Inside the Computer. After studying this chapter, students should be able to:

Data Storage. Chapter 3. Objectives. 3-1 Data Types. Data Inside the Computer. After studying this chapter, students should be able to: Chapter 3 Data Storage Objectives After studying this chapter, students should be able to: List five different data types used in a computer. Describe how integers are stored in a computer. Describe how

More information

GETTING STARTED WITH LABVIEW POINT-BY-POINT VIS

GETTING STARTED WITH LABVIEW POINT-BY-POINT VIS USER GUIDE GETTING STARTED WITH LABVIEW POINT-BY-POINT VIS Contents Using the LabVIEW Point-By-Point VI Libraries... 2 Initializing Point-By-Point VIs... 3 Frequently Asked Questions... 5 What Are the

More information

Low-resolution Character Recognition by Video-based Super-resolution

Low-resolution Character Recognition by Video-based Super-resolution 2009 10th International Conference on Document Analysis and Recognition Low-resolution Character Recognition by Video-based Super-resolution Ataru Ohkura 1, Daisuke Deguchi 1, Tomokazu Takahashi 2, Ichiro

More information

Implementing an In-Service, Non- Intrusive Measurement Device in Telecommunication Networks Using the TMS320C31

Implementing an In-Service, Non- Intrusive Measurement Device in Telecommunication Networks Using the TMS320C31 Disclaimer: This document was part of the First European DSP Education and Research Conference. It may have been written by someone whose native language is not English. TI assumes no liability for the

More information

REAL TIME TRAFFIC LIGHT CONTROL USING IMAGE PROCESSING

REAL TIME TRAFFIC LIGHT CONTROL USING IMAGE PROCESSING REAL TIME TRAFFIC LIGHT CONTROL USING IMAGE PROCESSING Ms.PALLAVI CHOUDEKAR Ajay Kumar Garg Engineering College, Department of electrical and electronics Ms.SAYANTI BANERJEE Ajay Kumar Garg Engineering

More information

Parallel Ray Tracing using MPI: A Dynamic Load-balancing Approach

Parallel Ray Tracing using MPI: A Dynamic Load-balancing Approach Parallel Ray Tracing using MPI: A Dynamic Load-balancing Approach S. M. Ashraful Kadir 1 and Tazrian Khan 2 1 Scientific Computing, Royal Institute of Technology (KTH), Stockholm, Sweden smakadir@csc.kth.se,

More information

High Quality Image Magnification using Cross-Scale Self-Similarity

High Quality Image Magnification using Cross-Scale Self-Similarity High Quality Image Magnification using Cross-Scale Self-Similarity André Gooßen 1, Arne Ehlers 1, Thomas Pralow 2, Rolf-Rainer Grigat 1 1 Vision Systems, Hamburg University of Technology, D-21079 Hamburg

More information

Instruction Set Architecture. or How to talk to computers if you aren t in Star Trek

Instruction Set Architecture. or How to talk to computers if you aren t in Star Trek Instruction Set Architecture or How to talk to computers if you aren t in Star Trek The Instruction Set Architecture Application Compiler Instr. Set Proc. Operating System I/O system Instruction Set Architecture

More information

MPSoC Designs: Driving Memory and Storage Management IP to Critical Importance

MPSoC Designs: Driving Memory and Storage Management IP to Critical Importance MPSoC Designs: Driving Storage Management IP to Critical Importance Design IP has become an essential part of SoC realization it is a powerful resource multiplier that allows SoC design teams to focus

More information

Analysis of an Artificial Hormone System (Extended abstract)

Analysis of an Artificial Hormone System (Extended abstract) c 2013. This is the author s version of the work. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purpose or for creating

More information

Generations of the computer. processors.

Generations of the computer. processors. . Piotr Gwizdała 1 Contents 1 st Generation 2 nd Generation 3 rd Generation 4 th Generation 5 th Generation 6 th Generation 7 th Generation 8 th Generation Dual Core generation Improves and actualizations

More information

Reliable Systolic Computing through Redundancy

Reliable Systolic Computing through Redundancy Reliable Systolic Computing through Redundancy Kunio Okuda 1, Siang Wun Song 1, and Marcos Tatsuo Yamamoto 1 Universidade de São Paulo, Brazil, {kunio,song,mty}@ime.usp.br, http://www.ime.usp.br/ song/

More information

Computer Graphics Hardware An Overview

Computer Graphics Hardware An Overview Computer Graphics Hardware An Overview Graphics System Monitor Input devices CPU/Memory GPU Raster Graphics System Raster: An array of picture elements Based on raster-scan TV technology The screen (and

More information

THE NAS KERNEL BENCHMARK PROGRAM

THE NAS KERNEL BENCHMARK PROGRAM THE NAS KERNEL BENCHMARK PROGRAM David H. Bailey and John T. Barton Numerical Aerodynamic Simulations Systems Division NASA Ames Research Center June 13, 1986 SUMMARY A benchmark test program that measures

More information

Outline. Introduction. Multiprocessor Systems on Chip. A MPSoC Example: Nexperia DVP. A New Paradigm: Network on Chip

Outline. Introduction. Multiprocessor Systems on Chip. A MPSoC Example: Nexperia DVP. A New Paradigm: Network on Chip Outline Modeling, simulation and optimization of Multi-Processor SoCs (MPSoCs) Università of Verona Dipartimento di Informatica MPSoCs: Multi-Processor Systems on Chip A simulation platform for a MPSoC

More information

Frequently Asked Questions

Frequently Asked Questions Frequently Asked Questions Minimum System Requirements What do I need to host or attend a meeting using Microsoft Windows? What do I need to host or attend a meeting using Mac OS? What do I need to host

More information

TriMedia CPU64 Application Development Environment

TriMedia CPU64 Application Development Environment Published at ICCD 1999, International Conference on Computer Design, October 10-13, 1999, Austin Texas, pp. 593-598. TriMedia CPU64 Application Development Environment E.J.D. Pol, B.J.M. Aarts, J.T.J.

More information

Force/position control of a robotic system for transcranial magnetic stimulation

Force/position control of a robotic system for transcranial magnetic stimulation Force/position control of a robotic system for transcranial magnetic stimulation W.N. Wan Zakaria School of Mechanical and System Engineering Newcastle University Abstract To develop a force control scheme

More information

Quiz for Chapter 1 Computer Abstractions and Technology 3.10

Quiz for Chapter 1 Computer Abstractions and Technology 3.10 Date: 3.10 Not all questions are of equal difficulty. Please review the entire quiz first and then budget your time carefully. Name: Course: Solutions in Red 1. [15 points] Consider two different implementations,

More information

Scalability and Classifications

Scalability and Classifications Scalability and Classifications 1 Types of Parallel Computers MIMD and SIMD classifications shared and distributed memory multicomputers distributed shared memory computers 2 Network Topologies static

More information

features at a glance

features at a glance hp availability stats and performance software network and system monitoring for hp NonStop servers a product description from hp features at a glance Online monitoring of object status and performance

More information

IMPROVING PERFORMANCE OF RANDOMIZED SIGNATURE SORT USING HASHING AND BITWISE OPERATORS

IMPROVING PERFORMANCE OF RANDOMIZED SIGNATURE SORT USING HASHING AND BITWISE OPERATORS Volume 2, No. 3, March 2011 Journal of Global Research in Computer Science RESEARCH PAPER Available Online at www.jgrcs.info IMPROVING PERFORMANCE OF RANDOMIZED SIGNATURE SORT USING HASHING AND BITWISE

More information

Cover. SEB SIMOTION Easy Basics. Collection of standardized SIMOTION basic functions. FAQ April 2011. Service & Support. Answers for industry.

Cover. SEB SIMOTION Easy Basics. Collection of standardized SIMOTION basic functions. FAQ April 2011. Service & Support. Answers for industry. Cover SEB SIMOTION Easy Basics Collection of standardized SIMOTION basic functions FAQ April 2011 Service & Support Answers for industry. 1 Preface 1 Preface The SEB is a collection of simple, standardized

More information

REMOTE RENDERING OF COMPUTER GAMES

REMOTE RENDERING OF COMPUTER GAMES REMOTE RENDERING OF COMPUTER GAMES Peter Eisert, Philipp Fechteler Fraunhofer Institute for Telecommunications, Einsteinufer 37, D-10587 Berlin, Germany eisert@hhi.fraunhofer.de, philipp.fechteler@hhi.fraunhofer.de

More information

Mean-Shift Tracking with Random Sampling

Mean-Shift Tracking with Random Sampling 1 Mean-Shift Tracking with Random Sampling Alex Po Leung, Shaogang Gong Department of Computer Science Queen Mary, University of London, London, E1 4NS Abstract In this work, boosting the efficiency of

More information

Introduction to Cloud Computing

Introduction to Cloud Computing Introduction to Cloud Computing Parallel Processing I 15 319, spring 2010 7 th Lecture, Feb 2 nd Majd F. Sakr Lecture Motivation Concurrency and why? Different flavors of parallel computing Get the basic

More information

Bindel, Spring 2010 Applications of Parallel Computers (CS 5220) Week 1: Wednesday, Jan 27

Bindel, Spring 2010 Applications of Parallel Computers (CS 5220) Week 1: Wednesday, Jan 27 Logistics Week 1: Wednesday, Jan 27 Because of overcrowding, we will be changing to a new room on Monday (Snee 1120). Accounts on the class cluster (crocus.csuglab.cornell.edu) will be available next week.

More information

Imaging and Bioinformatics Software

Imaging and Bioinformatics Software Imaging and Bioinformatics Software Software Overview 242 Gel Analysis Software 243 Ordering Information 246 Software Overview Software Overview See Also Imaging systems: pages 232 237. Bio-Plex Manager

More information

Detection and Restoration of Vertical Non-linear Scratches in Digitized Film Sequences

Detection and Restoration of Vertical Non-linear Scratches in Digitized Film Sequences Detection and Restoration of Vertical Non-linear Scratches in Digitized Film Sequences Byoung-moon You 1, Kyung-tack Jung 2, Sang-kook Kim 2, and Doo-sung Hwang 3 1 L&Y Vision Technologies, Inc., Daejeon,

More information

HANDS-FREE PC CONTROL CONTROLLING OF MOUSE CURSOR USING EYE MOVEMENT

HANDS-FREE PC CONTROL CONTROLLING OF MOUSE CURSOR USING EYE MOVEMENT International Journal of Scientific and Research Publications, Volume 2, Issue 4, April 2012 1 HANDS-FREE PC CONTROL CONTROLLING OF MOUSE CURSOR USING EYE MOVEMENT Akhil Gupta, Akash Rathi, Dr. Y. Radhika

More information

An Interactive Visualization Tool for the Analysis of Multi-Objective Embedded Systems Design Space Exploration

An Interactive Visualization Tool for the Analysis of Multi-Objective Embedded Systems Design Space Exploration An Interactive Visualization Tool for the Analysis of Multi-Objective Embedded Systems Design Space Exploration Toktam Taghavi, Andy D. Pimentel Computer Systems Architecture Group, Informatics Institute

More information

Synchronization of sampling in distributed signal processing systems

Synchronization of sampling in distributed signal processing systems Synchronization of sampling in distributed signal processing systems Károly Molnár, László Sujbert, Gábor Péceli Department of Measurement and Information Systems, Budapest University of Technology and

More information

Digitale Signalverarbeitung mit FPGA (DSF) Soft Core Prozessor NIOS II Stand Mai 2007. Jens Onno Krah

Digitale Signalverarbeitung mit FPGA (DSF) Soft Core Prozessor NIOS II Stand Mai 2007. Jens Onno Krah (DSF) Soft Core Prozessor NIOS II Stand Mai 2007 Jens Onno Krah Cologne University of Applied Sciences www.fh-koeln.de jens_onno.krah@fh-koeln.de NIOS II 1 1 What is Nios II? Altera s Second Generation

More information

Managing Relion IEDs Protection and control IED manager PCM600

Managing Relion IEDs Protection and control IED manager PCM600 Managing Relion IEDs Protection and control IED manager PCM600 Efficient management of Relion protection and control IEDs The protection and control IED manager PCM600 tool provides versatile functionalities

More information

Getting Started with Tableau Server 6.1

Getting Started with Tableau Server 6.1 Getting Started with Tableau Server 6.1 Welcome to Tableau Server. This guide will walk you through the basic steps to install and configure Tableau Server. Then follow along using sample data and users

More information

Achieving Nanosecond Latency Between Applications with IPC Shared Memory Messaging

Achieving Nanosecond Latency Between Applications with IPC Shared Memory Messaging Achieving Nanosecond Latency Between Applications with IPC Shared Memory Messaging In some markets and scenarios where competitive advantage is all about speed, speed is measured in micro- and even nano-seconds.

More information

EMC Publishing. Ontario Curriculum Computer and Information Science Grade 11

EMC Publishing. Ontario Curriculum Computer and Information Science Grade 11 EMC Publishing Ontario Curriculum Computer and Information Science Grade 11 Correlations for: An Introduction to Programming Using Microsoft Visual Basic 2005 Theory and Foundation Overall Expectations

More information

A Survey on ARM Cortex A Processors. Wei Wang Tanima Dey

A Survey on ARM Cortex A Processors. Wei Wang Tanima Dey A Survey on ARM Cortex A Processors Wei Wang Tanima Dey 1 Overview of ARM Processors Focusing on Cortex A9 & Cortex A15 ARM ships no processors but only IP cores For SoC integration Targeting markets:

More information

Grid Computing for Artificial Intelligence

Grid Computing for Artificial Intelligence Grid Computing for Artificial Intelligence J.M.P. van Waveren May 25th 2007 2007, Id Software, Inc. Abstract To show intelligent behavior in a First Person Shooter (FPS) game an Artificial Intelligence

More information

System Considerations

System Considerations System Considerations Interfacing Performance Power Size Ease-of Use Programming Interfacing Debugging Cost Device cost System cost Development cost Time to market Integration Peripherals Different Needs?

More information

Screen Capture A Vector Quantisation Approach

Screen Capture A Vector Quantisation Approach Screen Capture A Vector Quantisation Approach Jesse S. Jin and Sue R. Wu Biomedical and Multimedia Information Technology Group School of Information Technologies, F09 University of Sydney, NSW, 2006 {jesse,suewu}@it.usyd.edu.au

More information

TMS320C6000 Programmer s Guide

TMS320C6000 Programmer s Guide TMS320C6000 Programmer s Guide Literature Number: SPRU198K Revised: July 2011 Printed on Recycled Paper Preface Read This First About This Manual This manual is a reference for programming TMS320C6000

More information

Multimedia Systems Hardware & Software THETOPPERSWAY.COM

Multimedia Systems Hardware & Software THETOPPERSWAY.COM Multimedia Systems Hardware & Software THETOPPERSWAY.COM Table of Content 1. Categories of multimedia systems 2. Categories of multimedia devices 3. Evolution of multimedia PC 4. Authoring tools 5. Classification

More information

4.1 Introduction 4.2 Explain the purpose of an operating system 4.2.1 Describe characteristics of modern operating systems Control Hardware Access

4.1 Introduction 4.2 Explain the purpose of an operating system 4.2.1 Describe characteristics of modern operating systems Control Hardware Access 4.1 Introduction The operating system (OS) controls almost all functions on a computer. In this lecture, you will learn about the components, functions, and terminology related to the Windows 2000, Windows

More information

How To Fix Out Of Focus And Blur Images With A Dynamic Template Matching Algorithm

How To Fix Out Of Focus And Blur Images With A Dynamic Template Matching Algorithm IJSTE - International Journal of Science Technology & Engineering Volume 1 Issue 10 April 2015 ISSN (online): 2349-784X Image Estimation Algorithm for Out of Focus and Blur Images to Retrieve the Barcode

More information

Enabling Fingerprint Authentication in Embedded Systems for Wireless Applications

Enabling Fingerprint Authentication in Embedded Systems for Wireless Applications Abstract Enabling Authentication in Embedded Systems for Wireless Applications P.S. Cheng, Y.S. Moon, Z.G. Cao, K.C. Chan, T.Y. Tang In this paper, we study different methodologies for implementing fingerprint

More information

The Complete Performance Solution for Microsoft SQL Server

The Complete Performance Solution for Microsoft SQL Server The Complete Performance Solution for Microsoft SQL Server Powerful SSAS Performance Dashboard Innovative Workload and Bottleneck Profiling Capture of all Heavy MDX, XMLA and DMX Aggregation, Partition,

More information

Parametric Comparison of H.264 with Existing Video Standards

Parametric Comparison of H.264 with Existing Video Standards Parametric Comparison of H.264 with Existing Video Standards Sumit Bhardwaj Department of Electronics and Communication Engineering Amity School of Engineering, Noida, Uttar Pradesh,INDIA Jyoti Bhardwaj

More information

GEDAE TM - A Graphical Programming and Autocode Generation Tool for Signal Processor Applications

GEDAE TM - A Graphical Programming and Autocode Generation Tool for Signal Processor Applications GEDAE TM - A Graphical Programming and Autocode Generation Tool for Signal Processor Applications Harris Z. Zebrowitz Lockheed Martin Advanced Technology Laboratories 1 Federal Street Camden, NJ 08102

More information

Learning Outcomes. Simple CPU Operation and Buses. Composition of a CPU. A simple CPU design

Learning Outcomes. Simple CPU Operation and Buses. Composition of a CPU. A simple CPU design Learning Outcomes Simple CPU Operation and Buses Dr Eddie Edwards eddie.edwards@imperial.ac.uk At the end of this lecture you will Understand how a CPU might be put together Be able to name the basic components

More information

STUDY AND SIMULATION OF A DISTRIBUTED REAL-TIME FAULT-TOLERANCE WEB MONITORING SYSTEM

STUDY AND SIMULATION OF A DISTRIBUTED REAL-TIME FAULT-TOLERANCE WEB MONITORING SYSTEM STUDY AND SIMULATION OF A DISTRIBUTED REAL-TIME FAULT-TOLERANCE WEB MONITORING SYSTEM Albert M. K. Cheng, Shaohong Fang Department of Computer Science University of Houston Houston, TX, 77204, USA http://www.cs.uh.edu

More information

MONTE-CARLO SIMULATION OF AMERICAN OPTIONS WITH GPUS. Julien Demouth, NVIDIA

MONTE-CARLO SIMULATION OF AMERICAN OPTIONS WITH GPUS. Julien Demouth, NVIDIA MONTE-CARLO SIMULATION OF AMERICAN OPTIONS WITH GPUS Julien Demouth, NVIDIA STAC-A2 BENCHMARK STAC-A2 Benchmark Developed by banks Macro and micro, performance and accuracy Pricing and Greeks for American

More information

The continuum of data management techniques for explicitly managed systems

The continuum of data management techniques for explicitly managed systems The continuum of data management techniques for explicitly managed systems Svetozar Miucin, Craig Mustard Simon Fraser University MCES 2013. Montreal Introduction Explicitly Managed Memory systems lack

More information

Compute Cluster Server Lab 3: Debugging the parallel MPI programs in Microsoft Visual Studio 2005

Compute Cluster Server Lab 3: Debugging the parallel MPI programs in Microsoft Visual Studio 2005 Compute Cluster Server Lab 3: Debugging the parallel MPI programs in Microsoft Visual Studio 2005 Compute Cluster Server Lab 3: Debugging the parallel MPI programs in Microsoft Visual Studio 2005... 1

More information

A Study on M2M-based AR Multiple Objects Loading Technology using PPHT

A Study on M2M-based AR Multiple Objects Loading Technology using PPHT A Study on M2M-based AR Multiple Objects Loading Technology using PPHT Sungmo Jung, Seoksoo Kim * Department of Multimedia Hannam University 133, Ojeong-dong, Daedeok-gu, Daejeon-city Korea sungmoj@gmail.com,

More information

Guidelines for Software Development Efficiency on the TMS320C6000 VelociTI Architecture

Guidelines for Software Development Efficiency on the TMS320C6000 VelociTI Architecture Guidelines for Software Development Efficiency on the TMS320C6000 VelociTI Architecture WHITE PAPER: SPRA434 Authors: Marie Silverthorn Leon Adams Richard Scales Digital Signal Processing Solutions April

More information

GSM/GPRS PHYSICAL LAYER ON SANDBLASTER DSP

GSM/GPRS PHYSICAL LAYER ON SANDBLASTER DSP GSM/GPRS PHYSICAL LAYER ON SANDBLASTER DSP Raghunath Kalavai, Murugappan Senthilvelan, Sitij Agrawal, Sanjay Jinturkar, John Glossner Sandbridge Technologies, 1 North Lexington Avenue, White Plains, NY

More information

THE most significant value of software-defined radio

THE most significant value of software-defined radio A Fixed-Point DSP Architecture for Software-Defined Radio Wouter Kriegler and Gert-Jan van Rooyen Department of Electrical and Electronic Engineering University of Stellenbosch Abstract Software-defined

More information

How To Program With Adaptive Vision Studio

How To Program With Adaptive Vision Studio Studio 4 intuitive powerful adaptable software for machine vision engineers Introduction Adaptive Vision Studio Adaptive Vision Studio software is the most powerful graphical environment for machine vision

More information

ADVANCED PROCESSOR ARCHITECTURES AND MEMORY ORGANISATION Lesson-12: ARM

ADVANCED PROCESSOR ARCHITECTURES AND MEMORY ORGANISATION Lesson-12: ARM ADVANCED PROCESSOR ARCHITECTURES AND MEMORY ORGANISATION Lesson-12: ARM 1 The ARM architecture processors popular in Mobile phone systems 2 ARM Features ARM has 32-bit architecture but supports 16 bit

More information

ultra fast SOM using CUDA

ultra fast SOM using CUDA ultra fast SOM using CUDA SOM (Self-Organizing Map) is one of the most popular artificial neural network algorithms in the unsupervised learning category. Sijo Mathew Preetha Joy Sibi Rajendra Manoj A

More information

MMGD0203 Multimedia Design MMGD0203 MULTIMEDIA DESIGN. Chapter 3 Graphics and Animations

MMGD0203 Multimedia Design MMGD0203 MULTIMEDIA DESIGN. Chapter 3 Graphics and Animations MMGD0203 MULTIMEDIA DESIGN Chapter 3 Graphics and Animations 1 Topics: Definition of Graphics Why use Graphics? Graphics Categories Graphics Qualities File Formats Types of Graphics Graphic File Size Introduction

More information

A Tool for Generating Partition Schedules of Multiprocessor Systems

A Tool for Generating Partition Schedules of Multiprocessor Systems A Tool for Generating Partition Schedules of Multiprocessor Systems Hans-Joachim Goltz and Norbert Pieth Fraunhofer FIRST, Berlin, Germany {hans-joachim.goltz,nobert.pieth}@first.fraunhofer.de Abstract.

More information

Motion Activated Video Surveillance Using TI DSP

Motion Activated Video Surveillance Using TI DSP Motion Activated Video Surveillance Using TI DSP Ching-Kai Huang and Tsuhan Chen Dept. of Electrical and Computer Engineering Carnegie Mellon University 5000 Forbes Avenue, Pittsburgh, PA 15213 tsuhan@cmu.edu

More information

DUKANE Intelligent Assembly Solutions

DUKANE Intelligent Assembly Solutions PC Configuration Requirements: Configuration Requirements for ipc Operation The hardware and operating system of the PC must comply with a list of minimum requirements for proper operation with the ipc

More information

Eight Ways to Increase GPIB System Performance

Eight Ways to Increase GPIB System Performance Application Note 133 Eight Ways to Increase GPIB System Performance Amar Patel Introduction When building an automated measurement system, you can never have too much performance. Increasing performance

More information

AN IMPROVED DOUBLE CODING LOCAL BINARY PATTERN ALGORITHM FOR FACE RECOGNITION

AN IMPROVED DOUBLE CODING LOCAL BINARY PATTERN ALGORITHM FOR FACE RECOGNITION AN IMPROVED DOUBLE CODING LOCAL BINARY PATTERN ALGORITHM FOR FACE RECOGNITION Saurabh Asija 1, Rakesh Singh 2 1 Research Scholar (Computer Engineering Department), Punjabi University, Patiala. 2 Asst.

More information