Separation of image parts using 2-D parallel form recursive filters

Author

Sivaramakrishna, Radhika

Author_Institution

Dept. of Electr. & Comput. Eng., Manitoba Univ., Winnipeg, Man., Canada

Volume

5

Issue

1

fYear

1996

fDate

1/1/1996 12:00:00 AM

Firstpage

175

Lastpage

178

Abstract

This correspondence deals with a new technique to separate objects or image parts in a composite image. A parallel form extension of a 2-D Steiglitz-McBride method is applied to the discrete cosine transform (DCT) of the image containing the objects that are to be separated. The obtained parallel form is the sum of several filters or systems, where the impulse response of each filter corresponds to the DCT of one object in the original image. Preliminary results on an image with two objects show that the algorithm works well, even in the case where one object occludes another as well as in the case of moderate noise

Keywords

discrete cosine transforms; image processing; noise; recursive filters; transient response; 2D Steiglitz-McBride method; 2D parallel form recursive filters; composite image; discrete cosine transform; image parts separation; impulse response; moderate noise; parallel form extension; Biomedical signal processing; Discrete cosine transforms; Gabor filters; Image coding; Image processing; Image reconstruction; Neural networks; Optimized production technology; Signal processing algorithms; Spatial resolution;

fLanguage

English

Journal_Title

Image Processing, IEEE Transactions on

Publisher

ieee

ISSN

1057-7149

Type

jour

DOI

10.1109/83.481686

Filename

481686