Border delineation in image pyramids by concurrent tree growing

Jean Michel Jolion; Peter Meer; Azriel Rosenfeld

doi:10.1016/0167-8655(90)90120-Q

Border delineation in image pyramids by concurrent tree growing

Jean Michel Jolion, Peter Meer, Azriel Rosenfeld

School of Engineering, Electrical & Computer Engineering

Research output: Contribution to journal › Article › peer-review

3 Scopus citations

Abstract

A new method for delineation of compact objects in image pyramids is presented. The borders of the objects are detected in a low resolution representation of the input, a higher level of the pyramids of the pixels on the two sides of an edge are the roots for two classes (object and background). The two classes are employed in two independent top-down tree growing processes. The information is passed downward by adjusting confidence measures. The employment of multiple roots defined on the smoothed representation of the input contributes to the robustness of the method at very low signal-to-noise ratios.

Original language	English (US)
Pages (from-to)	107-115
Number of pages	9
Journal	Pattern Recognition Letters
Volume	11
Issue number	2
DOIs	https://doi.org/10.1016/0167-8655(90)90120-Q
State	Published - Feb 1990

All Science Journal Classification (ASJC) codes

Software
Signal Processing
Computer Vision and Pattern Recognition
Artificial Intelligence

Keywords

Image pyramids
feature extraction
multiresolution methods
noise smoothing

Access to Document

10.1016/0167-8655(90)90120-Q

Cite this

@article{73d6aab95a2f4565922af6a93b2100b6,

title = "Border delineation in image pyramids by concurrent tree growing",

abstract = "A new method for delineation of compact objects in image pyramids is presented. The borders of the objects are detected in a low resolution representation of the input, a higher level of the pyramids of the pixels on the two sides of an edge are the roots for two classes (object and background). The two classes are employed in two independent top-down tree growing processes. The information is passed downward by adjusting confidence measures. The employment of multiple roots defined on the smoothed representation of the input contributes to the robustness of the method at very low signal-to-noise ratios.",

keywords = "Image pyramids, feature extraction, multiresolution methods, noise smoothing",

author = "Jolion, {Jean Michel} and Peter Meer and Azriel Rosenfeld",

note = "Funding Information: Let an input image be of size 2{"} x 2t,. This image is placed at the base of the pyramid, level I = 0 . The next level, 1=L. of the pyramid is generated by smoothing the input with a weighted averaging window of size K x K . The window is usually called the kernel of the pyramid . Every second pixel in the horizontal and vertical directions is retained after smoothing . The retained pixels constitute the The support of the National Science foundation and the U .S . Air Force Office of Scientific Research under NSF Grant DCR 86-03723 is gratefully acknowledged .",

year = "1990",

month = feb,

doi = "10.1016/0167-8655(90)90120-Q",

language = "English (US)",

volume = "11",

pages = "107--115",

journal = "Pattern Recognition Letters",

issn = "0167-8655",

publisher = "Elsevier",

number = "2",

}

TY - JOUR

T1 - Border delineation in image pyramids by concurrent tree growing

AU - Jolion, Jean Michel

AU - Meer, Peter

AU - Rosenfeld, Azriel

N1 - Funding Information: Let an input image be of size 2" x 2t,. This image is placed at the base of the pyramid, level I = 0 . The next level, 1=L. of the pyramid is generated by smoothing the input with a weighted averaging window of size K x K . The window is usually called the kernel of the pyramid . Every second pixel in the horizontal and vertical directions is retained after smoothing . The retained pixels constitute the The support of the National Science foundation and the U .S . Air Force Office of Scientific Research under NSF Grant DCR 86-03723 is gratefully acknowledged .

PY - 1990/2

Y1 - 1990/2

N2 - A new method for delineation of compact objects in image pyramids is presented. The borders of the objects are detected in a low resolution representation of the input, a higher level of the pyramids of the pixels on the two sides of an edge are the roots for two classes (object and background). The two classes are employed in two independent top-down tree growing processes. The information is passed downward by adjusting confidence measures. The employment of multiple roots defined on the smoothed representation of the input contributes to the robustness of the method at very low signal-to-noise ratios.

AB - A new method for delineation of compact objects in image pyramids is presented. The borders of the objects are detected in a low resolution representation of the input, a higher level of the pyramids of the pixels on the two sides of an edge are the roots for two classes (object and background). The two classes are employed in two independent top-down tree growing processes. The information is passed downward by adjusting confidence measures. The employment of multiple roots defined on the smoothed representation of the input contributes to the robustness of the method at very low signal-to-noise ratios.

KW - Image pyramids

KW - feature extraction

KW - multiresolution methods

KW - noise smoothing

UR - http://www.scopus.com/inward/record.url?scp=45149137865&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=45149137865&partnerID=8YFLogxK

U2 - 10.1016/0167-8655(90)90120-Q

DO - 10.1016/0167-8655(90)90120-Q

M3 - Article

AN - SCOPUS:45149137865

VL - 11

SP - 107

EP - 115

JO - Pattern Recognition Letters

JF - Pattern Recognition Letters

SN - 0167-8655

IS - 2

ER -

Border delineation in image pyramids by concurrent tree growing

Abstract

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this