CLC number: TP391.41
On-line Access:
Received: 1999-04-06
Revision Accepted: 2000-01-08
Crosschecked: 0000-00-00
Cited: 0
Clicked: 5804
WANG Hong-yu, LI Hong-dong, YE Xiu-qing, GU Wei-kang. TRAINING A NEURAL NETWORK FOR MOMENT BASED IMAGE EDGE DETECTION[J]. Journal of Zhejiang University Science A, 2000, 1(4): 398-401.
@article{title="TRAINING A NEURAL NETWORK FOR MOMENT BASED IMAGE EDGE DETECTION",
author="WANG Hong-yu, LI Hong-dong, YE Xiu-qing, GU Wei-kang",
journal="Journal of Zhejiang University Science A",
volume="1",
number="4",
pages="398-401",
year="2000",
publisher="Zhejiang University Press & Springer",
doi="10.1631/jzus.2000.0398"
}
%0 Journal Article
%T TRAINING A NEURAL NETWORK FOR MOMENT BASED IMAGE EDGE DETECTION
%A WANG Hong-yu
%A LI Hong-dong
%A YE Xiu-qing
%A GU Wei-kang
%J Journal of Zhejiang University SCIENCE A
%V 1
%N 4
%P 398-401
%@ 1869-1951
%D 2000
%I Zhejiang University Press & Springer
%DOI 10.1631/jzus.2000.0398
TY - JOUR
T1 - TRAINING A NEURAL NETWORK FOR MOMENT BASED IMAGE EDGE DETECTION
A1 - WANG Hong-yu
A1 - LI Hong-dong
A1 - YE Xiu-qing
A1 - GU Wei-kang
J0 - Journal of Zhejiang University Science A
VL - 1
IS - 4
SP - 398
EP - 401
%@ 1869-1951
Y1 - 2000
PB - Zhejiang University Press & Springer
ER -
DOI - 10.1631/jzus.2000.0398
Abstract: edge detection is a crucial step to computer vision. Currently, there is not a single edge detector that has both efficiency and reliability. Traditional differential filter-based algorithms have the advantage of theoretical strictness , but require excessive post-processing. This paper introduces a neural network edge detector that takes advantage of moments features. It functions as a neural pattern classifier that directly estimates the posterior probability from the training data set. Two subsystems can be distinguished and different kinds of learning rules are used. For the end-user, it works as a black box that directly transforms raw images into the edge maps so no complicated postprocessing is required. Tests on both simulated and real images showed the proposed neural network edge detector is superior to traditional operators.
[1]Canny, J., 1986. A computational approach to edge detection. IEEE Trans. on Pattern Analysis and Machine Intelligence, 8 (1): 679-698.
[2]Kosko, B.,1991. Neural Networks and Fuzzy Systems: A Dynamics System to Machine Intelligence. Printice-Hall, New York.
[3]Lyvers, E.,1988. Precision edge contrast and orientation estimation,IEEE Trans. on Pattern Analysis and Machine Intelligence, 10: 927-937.
[4]Lyvers, E., 1989. Subpixel measurements using a moment based edge operator, IEEE Trans. on Pattern Analysis and Machine Intelligence, 11(12): 63-72.
[5]Machuca, R.,1981. Finding edges in noisy scenes, IEEE Trans. on Pattern Analysis and Machine Intelligence, 3: 103-111.
[6]Marr, D., Hildreth, E., 1980. Theory of edge detection, Proc. of Royal Society Landon, 1980, B(207): 187-217.
[7]Pinho, A.J., Almeida, L.B., 1995. Edge detection filters based on artificial neural networks, Pro. of ICIAP'95, IEEE Computer Society Press, p.159-164.
[8]Speeruses, L.J., 1994. Neural network edge detector, SPIE ,2451: 204-215.
Open peer comments: Debate/Discuss/Question/Opinion
<1>