High Productivity Computing System Program

1
Wavelet Spectral Dimension Reduction of
Hyperspectral Imagery on a Reconfigurable
ComputerTarek El-Ghazawi1, Esam El-Araby1,
Abhishek Agarwal1, Jacqueline Le Moigne2, and
Kris Gaj31The George Washington
University,2NASA/Goddard Space Flight
Center,3George Mason Universitytarek, esam,
agarwala_at_gwu.edu, lemoigne_at_backserv.gsfc.nasa.gov
, kgaj_at_gmu.edu
2
Objectives and Introduction

• Investigate Use of Reconfigurable Computing for
• On-Board Automatic Processing of
• Remote Sensing Data
• Remote Sensing ? Image Classification
• Applications
• Land Classification, Mining, Geology, Forestry,
  Agriculture, Environmental Management, Global
  Atmospheric Profiling (e.g. water vapor and
  temperature profiles), and Planetary Space
  missions
• Types of Carriers

3
Types of Sensing

• Mono-Spectral Imagery ? 1 band (SPOT
  panchromatic)
• Multi-Spectral Imagery ? 10s of bands (MODIS 36
  bands, SeaWiFS 8 bands, IKONOS 5 bands)
• Hyperspectral Imagery ? 100s-1000s of bands
  (AVIRIS 224 bands, AIRS 2378 bands)

4
Different Airborne Hyperspectral Systems
5
Why On-Board Processing?

• Problems
• Complex Pre-processing Steps
• Image Registration / Fusion
• Large Data Volumes
• Large cost and complexity of the On-The-Ground /
  Earth processing systems
• Large critical decisions latency
• Large data downlink bandwidth requirements

• Solutions
• Automatic On-Board Processing
• Reduces the cost and the complexity of the
  On-The-Ground/Earth processing system
• larger utilization for broader community,
  including educational institutions
• Enables autonomous decisions to be taken on-board
  ? faster critical decisions
• Applications
• Future reconfigurable web sensors missions
• Future Mars and planetary exploration missions
• Dimension Reduction
• Reduction of communication bandwidth
• Simpler and faster subsequent computations

Investigated Pre-Processing Step
6
Why Reconfigurable Computers?

• On-Board Processing Problems
• High Computational Complexities
• Low performance for traditional processing
  platforms
• High form / wrap factors (size and weight) for
  parallel computing systems
• Low flexibility for traditional ASIC-Based
  solutions
• High costs and long design cycles for traditional
  ASIC-Based solutions

• Solutions
• Reconfigurable Computers (RCs)
• Higher performance (throughput and processing
  power) compared to conventional processors
• Lower form / wrap factors compared to parallel
  computers
• Higher flexibility (reconfigurability) compared
  to ASICs
• Less costs and shorter time-to-solution compared
  to ASICs

7
Introduction
8
Data Arrangement
9
Data Arrangement (cntd)
Pixels Rows X Columns
10
Examples of Hyperspectral Datasets
11
Dimension Reduction Techniques

• Principal Component Analysis (PCA)
• Most Common Method Dimension Reduction
• Does Not Preserve Spectral Signatures
• Complex and Global computations difficult for
  parallel processing and hardware implementations
• Wavelet-Based Dimension Reduction
• Preserves Spectral Signatures
• High-Performance Implementation
• Simple and Local Operations

12
2-D DWT (1-level Decomposition)
13
2-D DWT (2-level Decomposition)
14
Wavelet-Based vs. PCA (Execution Time, 500 MHz
P3)
Complexity Wavelet-Based O(MN) PCA
O(MN2N3)
15
Wavelet-Based vs. PCA (cntd) (Execution Time,
500 MHz P3)
Complexity Wavelet-Based O(MN) PCA
O(MN2N3)
16
Wavelet-Based vs. PCA (cntd) (Classification
Accuracy)

• Implemented on the HIVE (8 Pentium
  Xeon/Beowulfs-Type System) 6.5 times faster than
  sequential implementation
• Classification Accuracy Similar or Better than
  PCA
• Faster than PCA

17
The Algorithm
18
Prototyping Wavelet-Based Dimension Reduction of
Hyperspectral Imagery on a Reconfigurable
Computer, the SRC-6E
19
Hardware Architecture of SRC-6E
20
SRC Compilation Process
21
Top Hierarchy Module
22
Decomposition and Reconstruction Levels of
Dimension Reduction (DWT_IDWT)
23
FIR Filters (L, L) Implementation
24
Correlator Module
25
Histogram Module
26
Resource Utilization and Operating Frequency
27
Measurements Scenarios
28
SRC Experiment Setup and Results

• Salinas98
• 217 X 512 Pixels, 192 Bands 162.75 MB
• Number of Streams 41
• Stream Size 2730 voxels 4 MB
• Non-Overlapped Streams
• TDMA-IN 13.040 msec
• TCOMP 0.62428 msec
• TDMA-OUT 22.712 msec
• TTotal 1.49 sec
• Throughput 109.23 MB/Sec
• Overlapped Streams
• TDMA 35.752 msec
• TCOMP 0.62428 msec
• Xc 0.0175

29
Execution Time
30
Distribution of Execution Times
31
Speedup Results
32
Concluding Remarks

• We prototyped the automatic wavelet-based
  dimension reduction algorithm on a reconfigurable
  architecture
• Both coarse-grain and fine-grain parallelism are
  exploited
• We observed a 10x speedup using the P3 version of
  SRC-6E. From our previous experience we expect
  this speedup to double using the P4 version of
  SRC machine
• These speedup figures were obtained while I/O is
  still dominating. The speedup can be increased by
  improving I/O Bandwidth of the reconfigurable
  platforms