CLONER
Project Director:
Leah H. Jamieson, School of Electrical Engineering, Purdue University
Project Participants:
Jamshed Patel, Pramila Srinivasan, Ph.D. Students
Ashfaq Khokhar, Visiting Research Assistant Professor
Cloner is a prototyping environment for computer vision and image
processing (CVIP) algorithms and tasks. It is being designed to
allow users to take advantage of the computing power provided by
parallel processing systems without requiring an extensive
understanding of the underlying architecture.
OBJECTIVES
- Create a polyalgorithmic library of low-, mid-, and high-level
computer vision and image processing algorithms. These
libraries are optimized for the target architectures.
- Provide a high-level visual environment that encapsulates
machine-specific expertise, as well as algorithm and
architecture-related scalability information.
- Provide the means for user-directed code modification to
implement algorithm prototyping.
PROJECT OVERVIEW
Cloner Visual Environment:
- Supports menu-based and query-based interaction with
the user via a graphical
X-window-based interface.
- Encapsulates machine-specific expertise, as well as
algorithm and architecture-related scalability information.
- Provides the means for user-directed code modification to
implement algorithm prototyping.
Computer Vision and Image Processing Library:
- Polyalgorithmic : algorithm classes are represented by multiple
algorithms and implementations to achieve scalability
- Common Data Structures
- common program interface within a level of the CVIP hierarchy
- flexible program interface across levels
Algorithm Prototyping:
- CVIP algorithms exhibit regular structure and recurring patterns.
Cloner exploits these properties of the problem domain to
provide a framework for software reuse.
- Code decomposition via annotation of library code:
- control structures for the execution of the parallel
program
- components that specify the image processing functionality
- New algorithms inherit implementation details and mapping
information from existing library templates
PROJECT STATUS
- Completing library of low-level image processing algorithms.
- Conducting performance evaluation studies.