Image Compression System Megan Fuller and Ezzeldin Hamed 1.

Slides:



Advertisements
Similar presentations
DFT & FFT Computation.
Advertisements

[1] AN ANALYSIS OF DIGITAL WATERMARKING IN FREQUENCY DOMAIN.
Fourier Transforms and Their Use in Data Compression
Processor Architecture Needed to handle FFT algoarithm M. Smith.
ECE 734: Project Presentation Pankhuri May 8, 2013 Pankhuri May 8, point FFT Algorithm for OFDM Applications using 8-point DFT processor (radix-8)
Fourier Transform (Chapter 4)
CENG536 Computer Engineering Department Çankaya University.
A Matlab Playground for JPEG Andy Pekarske Nikolay Kolev.
ELEC692 VLSI Signal Processing Architecture Lecture 9 VLSI Architecture for Discrete Cosine Transform.
Department of Computer Engineering University of California at Santa Cruz Data Compression (3) Hai Tao.
Sampling, Reconstruction, and Elementary Digital Filters R.C. Maher ECEN4002/5002 DSP Laboratory Spring 2002.
Digital Image Processing Chapter 4: Image Enhancement in the Frequency Domain.
Introduction to Fast Fourier Transform (FFT) Algorithms R.C. Maher ECEN4002/5002 DSP Laboratory Spring 2003.
Chapter 15 Digital Signal Processing
Image Compression and Signal Processing Dan Hewett CS 525.
CS :: Fall 2003 MPEG-1 Video (Part 1) Ketan Mayer-Patel.
T.Sharon-A.Frank 1 Multimedia Image Compression 2 T.Sharon-A.Frank Coding Techniques – Hybrid.
Digital Image Processing Final Project Compression Using DFT, DCT, Hadamard and SVD Transforms Zvi Devir and Assaf Eden.
Input image Output image Transform equation All pixels Transform equation.
CELLULAR COMMUNICATIONS DSP Intro. Signals: quantization and sampling.
Fast Fourier Transforms
Image Compression - JPEG. Video Compression MPEG –Audio compression Lossy / perceptually lossless / lossless 3 layers Models based on speech generation.
Prepared by: Hind J. Zourob Heba M. Matter Supervisor: Dr. Hatem El-Aydi Faculty Of Engineering Communications & Control Engineering.
JPEG C OMPRESSION A LGORITHM I N CUDA Group Members: Pranit Patel Manisha Tatikonda Jeff Wong Jarek Marczewski Date: April 14, 2009.
ECE472/572 - Lecture 12 Image Compression – Lossy Compression Techniques 11/10/11.
Processor Architecture Needed to handle FFT algoarithm M. Smith.
8.1 representation of periodic sequences:the discrete fourier series 8.2 the fourier transform of periodic signals 8.3 properties of the discrete fourier.
Transforms. 5*sin (2  4t) Amplitude = 5 Frequency = 4 Hz seconds A sine wave.
Digital Signal Processing – Chapter 10
Fast Memory Addressing Scheme for Radix-4 FFT Implementation Presented by Cheng-Chien Wu, Master Student of CSIE,CCU 1 Author: Xin Xiao, Erdal Oruklu and.
J. Greg Nash ICNC 2014 High-Throughput Programmable Systolic Array FFT Architecture and FPGA Implementations J. Greg.
Part I: Image Transforms DIGITAL IMAGE PROCESSING.
Zhongguo Liu_Biomedical Engineering_Shandong Univ. Chapter 8 The Discrete Fourier Transform Zhongguo Liu Biomedical Engineering School of Control.
Z TRANSFORM AND DFT Z Transform
7- 1 Chapter 7: Fourier Analysis Fourier analysis = Series + Transform ◎ Fourier Series -- A periodic (T) function f(x) can be written as the sum of sines.
Digital Signal Processing
Copyright © 2004, Dillon Engineering Inc. All Rights Reserved. An Efficient Architecture for Ultra Long FFTs in FPGAs and ASICs  Architecture optimized.
Fourier Transform.
Chapter 8 Lossy Compression Algorithms. Fundamentals of Multimedia, Chapter Introduction Lossless compression algorithms do not deliver compression.
A New Class of High Performance FFTs Dr. J. Greg Nash Centar ( High Performance Embedded Computing (HPEC) Workshop.
Fast Fourier Transforms. 2 Discrete Fourier Transform The DFT pair was given as Baseline for computational complexity: –Each DFT coefficient requires.
Chapter 13 Discrete Image Transforms
An FFT for Wireless Protocols Dr. J. Greg Nash Centar ( HAWAI'I INTERNATIONAL CONFERENCE ON SYSTEM SCIENCES Mobile.
Fourier transform.
Signal Prediction and Transformation Trac D. Tran ECE Department The Johns Hopkins University Baltimore MD
Discrete Fourier Transform
1 Paper reading A New Approach to FFT Processor Speaker: 吳紋浩 第六組 洪聖揚 吳紋浩 Adviser: Prof. Andy Wu Mentor: 陳圓覺.
1 Chapter 8 The Discrete Fourier Transform (cont.)
Digital Image Processing Lecture 8: Fourier Transform Prof. Charlene Tsai.
Chapter 8 Lossy Compression Algorithms
Singular Value Decomposition and its applications
JPEG Compression What is JPEG? Motivation
Last update on June 15, 2010 Doug Young Suh
Even Discrete Cosine Transform The Chinese University of Hong Kong
Fourier Transform.
Centar ( Global Signal Processing Expo
Lecture #17 INTRODUCTION TO THE FAST FOURIER TRANSFORM ALGORITHM
JPEG Image Compression
Chapter 8 The Discrete Fourier Transform
Z TRANSFORM AND DFT Z Transform
JPEG Still Image Data Compression Standard
1-D DISCRETE COSINE TRANSFORM DCT
DCT-based Processing of Dynamic Features for Robust Speech Recognition Wen-Chi LIN, Hao-Teng FAN, Jeih-Weih HUNG Wen-Yi Chu Department of Computer Science.
C Model Sim (Fixed-Point) -A New Approach to Pipeline FFT Processor
Chapter 8 The Discrete Fourier Transform
Digital Image Procesing Discrete CosineTrasform (DCT) in Image Processing DR TANIA STATHAKI READER (ASSOCIATE PROFFESOR) IN SIGNAL PROCESSING IMPERIAL.
Chapter 8 The Discrete Fourier Transform
Introduction SYSC5603 (ELG6163) Digital Signal Processing Microprocessors, Software and Applications Miodrag Bolic.
Introduction SYSC5603 (ELG6163) Digital Signal Processing Microprocessors, Software and Applications Miodrag Bolic.
Speaker: Chris Chen Advisor: Prof. An-Yeu Wu Date: 2014/10/28
Presentation transcript:

Image Compression System Megan Fuller and Ezzeldin Hamed 1

Transforms of Images Original Image Image Reconstructed from 25% of DFT coefficients Magnitude of DFT of Image-128 (otherwise DC component = ~8e6) 2

The 2D Discrete Fourier Transform 3

The 2D Discrete Cosine Transform 4

High Level Architecture Separable, in- place 2D DFT/DCT Input Memory Coefficient > Threshold? Output Module (sending data to PC) 5 The choice between DFT and DCT is provided at compile time Threshold is provided by the user at run time

What’s Interesting? Reducing the computation required Sharing resources in the DCT case Some memory organization tricks Reducing bit width 6

Number of FFTs 7

Reduction for the DFT case 8 S 00 S 01 S 02 S 03 S 10 S 11 S 12 S 13 S 20 S 21 S 22 S 23 S 30 S 31 S 32 S 33 N/2 FFTs of the rows, followed by Even/Odd decomposition Output is symmetric (discard half the columns) N/2 FFTs of the columns Total of N FFT computations S 31 S 11 RealImag

Reduction in the DCT case Again combining the rows in the same way as in DFT (N/2 FFTs) Even/Odd decomposition then extra multiplication to calculate the DCT 9 S 10 S 11 S 12 S 13 S 00 S 01 S 02 S 03 S 30 S 31 S 32 S 33 S 20 S 21 S 22 S 23 Results are not symmetric But the DCT is real We can combine the columns the same way we combined the rows (N/2 FFT) The same multiplier inside the FFT is used Another Even/Odd decomposition is required here with an extra complex multiplier Total of N FFT computations + few extra multiplications RealImag

In-Place Radix-4 FFT 10

Static Scaling Vs. Dynamic Scaling Shift when you expect an overflow – Shift after each addition The location of the fraction point is fixed at each computation step Almost no overhead compared to fixed point Higher effective bit width only in the first computation steps No effect on the critical path 11 Shift only when overflow occurs – Track overflows and account for them The location of the fraction point is the same for each 1D-FFT frame Needs simple circuitry to track the overflow and shift when required Effective bit width depend on the data. No effect on the critical path

Design Space Explored Dynamic Scaling YesNo DFTDCTDFT DCT bits with dynamic scaling considered later 8 bits without dynamic scaling (and 12 for DCT) perform too poorly to be considered 12 does as good as 16 bits with dynamic scaling in the DFT

Dynamic Scaling of DFT 13 50% of coefficients is sufficient for perfect reconstruction because of the symmetry of the DFT 16 bits without dynamic scaling does as well as floating point 12 bits with dynamic scaling also does nearly as well as floating point

Dynamic Scaling of DFT(continued) 14 Improvement in performance when dynamic scaling is used more than makes up for reduced compression because the scaling bits have to be saved 12 bits with dynamic scaling does nearly as well as 16 bits

DCT Vs. DFT 15 All cases are using dynamic scaling DCT provides better energy compaction For DCT, 12 bits gives a lower MSE for a given compression ratio (this was not the case for the DFT).

8 Bits Image reconstructed from 50% of the DFT coefficients, computed with 8 bits, using dynamic scaling. MSE = 452. Image reconstructed from 6% of the DFT coefficients, computed with 16 bits, MSE =

Physical Considerations Transform# of BitsDynamic Scaling? Critical PathSlice Registers Slice LUTs BRAMDSP48Es DFT16No11.458ns16%23%29%7 DFT16Yes11.763ns17%24%29%7 DFT12No11.273ns15%22%24%7 DFT12Yes11.464ns16%23%24%7 DFT8Yes11.287ns15%22%18%6 DCT16Yes11.458ns19%26%29%10 DCT12Yes11.273ns18%25%24%10 DCT8Yes11.066ns17%23%18%8 17 Critical path about the same for all designs, could probably be improved with tighter synthesis constraints Resource usage increases with bitwidth, addition of dynamic scaling, and DCT, but overall doesn’t change much DCT uses extra DSP blocks because of the extra multiplication

Latency ComponentLatency (clock cycles)Potential Frame Rate with 50MHz Clock Initialization870,000- DCT263, images/second DFT262, images/second 18

Future Work Use of DRAM to allow compression of larger images Support for color images Support for rectangular images of arbitrary edge length Combining the DCT and DFT into a single core that could compute either transform, as selected by the user at runtime 19

Relationship Between the DFT and the DCT The N-point DFT of a sequence is the Fourier Series coefficients for that sequence made periodic with period N. 20

Relationship Between the DFT and the DCT (continued) The N-point DCT of a sequence is a twiddle factor multiplied by the first N Fourier Series coefficients of the 2N point sequence y(n) made periodic with period 2N. y(n) = x(x) + x(2N-1-n) x(n) 21

Relationship Between the DFT and the DCT (continued) 22

Rounding DesignMSE Decrease with Rounding 12 bits, no dynamic scaling, DFT20 16 bits, no dynamic scaling, DFT0 12 bits, dynamic scaling, DFT2 16 bits, no dynamic scaling, DCT0 12 bits, dynamic scaling, DCT2 16 bits, dynamic scaling, DCT0 Conclusion: Never hurt, often helped. Free in hardware (just a register initialization), so always use it. All subsequent results will be using rounding. 23

Dynamic Scaling of DCT 24

Dynamic Scaling of DCT (continued) 25

Limitations of MSE Image reconstructed from 5.7% of the DCT coefficients, computed with dynamic scaling. MSE = 193 Image reconstructed from 6.1% of the DCT coefficients, computed without dynamic scaling. MSE =

Performance of 8 Bit Systems 27

More Limitations of MSE (Left) 8 bit DFT coefficients, computed with rounding. Compression ratio = 2.3, MSE = 869. (Right) 8 bit DFT coefficients, computed without rounding. Compression ratio = 2.1, MSE = 664 (Left) 8 bit DCT coefficients, computed with rounding. Compression ratio = 2.2, MSE = 517. (Right) 8 bit DCT coefficients, computed without rounding. Compression ratio = 2.4, MSE =