University of Washington Image Computing Library (UWICL)
by Image Computing Systems Laboratory of the University of Washington

Software Overview

We have developed a highly-optimized image-computing library for the TI TMS320C80. Our goals in 
developing UWICL are to: 1) provide an efficient and flexible library of core low-level algorithms 
to the 'C80 user community, 2) provide a portable infrastructure of `C80 routines so that 
higher-level algorithms and applications can be developed and quickly integrated on the `C80-based 
target system with minimal custom coding, 3) actively expand, upgrade, and support the library to 
facilitate the widespread use of the `C80, and 
4) educate students in image-computing algorithms and `C80 programming.
The UWICL functions are divided into three hierarchical segments, i.e., MP-level API, DSP-level 
API, and API for processing cores. The MP-level API modules interact with the host processor and 
`C80 ADSPs passing commands and parameters to ADSPs. The DSP-level API modules set packet transfers 
and call ADSP tight-loop functions. The processing core modules process all on-chip data according 
to imaging algorithms and are coded in `C80 ADSP assembly language. The UWICL Design Guide 
documents this architecture in detail while the UWICL Style Guides specify the coding standards. 
Currently, the library has 94 routines ranging from convolution and morphology to wavelets, FFT, 
rotation, and warping, and more functions are being added. With UWICL, fast implementation using 
the library routines as building blocks can be realized. Programmers do not need to understand the 
details of the `C80 ADSP architecture to be able to maximize their processing power.

Features and Benefits

•	Highly optimized, thus providing high performance
•	Standardized design and styles with good documentation
•	Flexible via multi-level hierarchical APIs
•	University of Washington's innovation in imaging, experience in `C80 programming and `C80-based 
systems, and independence and stability.
•	Fast implementation of a high-level algorithm or an application

Processor and System Requirements

•	Devices supported: TMS320C80, TMS320C82
•	Algorithm category: Image

Algorithm Verification

•	Quality assurance program with white-box and black-box testing using test images is applied to 
every routine before an official software release.

Availability

•	April 1995 and updated twice a year in April and October through the University of Washington 
Image Computing Library Consortium.

Company Background and Contact Information

The Image Computing Systems Laboratory of the University of Washington participated in the `C80 
architecture definition and optimization as well as algorithm simulation and system development 
since 1990. The laboratory has been performing research and development in multimedia, image 
processing and computer graphics, medical imaging, advanced processor and workstation design, ASIC 
design, PACS, modeling, and simulation.
Dr. Kim and his research group have made many inventions in imaging and multimedia systems, and 
transferred the invented technologies to industry with 10 licenses and helped commercialization of 
these technologies.
 
Contact:	Dr. Yongmin Kim, Ph.D.
		Professor and Director of the Image Computing Systems Laboratory
		Department of Electrical Engineering
		University of Washington
		P.O. Box 352500
		Seattle, WA 98195-2500  USA
		(206) 685-2271
		Fax: (206) 543-3842
		e-mail: kim@ee.washington.edu
		www: icsl.ee.washington.edu/projects/iclib/