Lab View  
HOME PEOPLE RESEARCH PUBLICATIONS COURSES SPONSORS    

RESEARCH  HIGHLIGHTS

     Projects

     Demos

     Software

     Datasets

  ABOUT US

     Seminars

     Lab

     News

 

  LINKS

Igert

Vision

Ece

Ucsb

Melvyl

 

 

 

 

Project: Color Image Filtering

PEOPLE

Canhui Cai, Luca Lucchese, S.K.Mitra

OBJECTIVE

Two specific problems were considered under this project.  The first problem was on the accurate determination of edges in a color image based on a quaternion rep­resentation, while the second problem was filtering of color images based the Center of Gravity Law of Color Mixture

A quaternion number is a generalization of the con­ventional complex number.  It consists of a real part and three imaginary parts.  An RGB color image can be rep­resented by a quaternion sys­tem of pixels in which the three imaginary parts represent the red, green, and the blue components, while the real part is set to zero.  A normalized color difference edge detector has been devel­oped based on this new type of rep­resentation.  The edge detector generates edges only where sharp changes of color occur in the origi­nal color image.  Experimental re­sults have verified the improved performance of the new edge detector compared to some well-known methods.  A paper describing the new edge detector along with the ex­per­imental results was presented at an international conference [1].

We have developed a new approach to color filtering which operates on the representation of an im­age in the xyY color space.  In this color space the chromatic in­formation is carried by the CIE chromatic coordinates, x and y , and the in­tensity informa­tion by the CIE tristimulus value Y .  The particular struc­ture of its xy chromatic subspace requires one to add colors in compliance with the Center of Gravity Law of Color Mixture .  Our origi­nal contribution consists in de­riving from this law a simple color filtering scheme, nonlinear in the x and y components, and linear in Y , which preserves the convexity of the chromatic informa­tion.  This impor­tant property has been demon­strated by a number of practical filtering examples.  A paper con­taining the initial results was presented at an international conference [2].

PUBLICATIONS

These materials are presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each authors copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.

 

 

If you have any comments or questions, please send them to web administrator.

Image Processing and Vision Research Lab, Electrical and Computer Engineering 

Department,  University of California at Santa Barbara, Santa Barbara, CA 93106.