Efficient edge detection in digital images using a cellular neural network optimized by differential evolution algorithm

被引:58
作者
Bastuerk, Alper [1 ]
Guenay, Enis [2 ]
机构
[1] Erciyes Univ, Dept Comp Engn, TR-38039 Kayseri, Turkey
[2] Erciyes Univ, Dept Elect & Elect Engn, TR-38039 Kayseri, Turkey
关键词
Cellular neural networks; Cloning template; Differential evolution algorithm; Edge detection; TEMPLATE; NOISY;
D O I
10.1016/j.eswa.2008.01.082
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
A cellular neural network (CNN) based edge detector optimized by differential evolution (DE) algorithm is presented. Cloning template of the proposed CNN is adaptively tuned by using simple training images. The performance of the proposed edge detector is evaluated on different test images and compared with popular edge detectors from the literature. Simulation results indicate that the proposed CNN operator outperforms competing edge detectors and offers Superior performance in edge detection in digital images. (C) 2008 Elsevier Ltd. All rights reserved.
引用
收藏
页码:2645 / 2650
页数:6
相关论文
共 39 条
[31]   NEIGHBORHOOD CODING OF BINARY IMAGES FOR FAST CONTOUR FOLLOWING AND GENERAL BINARY ARRAY PROCESSING [J].
SOBEL, I .
COMPUTER GRAPHICS AND IMAGE PROCESSING, 1978, 8 (01) :127-135
[32]   EDGE-DETECTION IN CORRELATED NOISE USING LATIN SQUARE MASKS [J].
STERN, D ;
KURZ, L .
PATTERN RECOGNITION, 1988, 21 (02) :119-129
[33]   Differential evolution - A simple and efficient heuristic for global optimization over continuous spaces [J].
Storn, R ;
Price, K .
JOURNAL OF GLOBAL OPTIMIZATION, 1997, 11 (04) :341-359
[34]  
*USC SIPI, TEXT DAT
[35]   Edge detection in noisy images by neuro-fuzzy processing [J].
Yuksel, M. Emin .
AEU-INTERNATIONAL JOURNAL OF ELECTRONICS AND COMMUNICATIONS, 2007, 61 (02) :82-89
[36]   A simple neuro-fuzzy edge detector for digital images corrupted by impulse noise [J].
Yüksel, ME ;
Yildirim, MT .
AEU-INTERNATIONAL JOURNAL OF ELECTRONICS AND COMMUNICATIONS, 2004, 58 (01) :72-75
[37]  
Zarándy A, 1999, INT J CIRC THEOR APP, V27, P5, DOI 10.1002/(SICI)1097-007X(199901/02)27:1<5::AID-CTA38>3.0.CO
[38]  
2-C
[39]  
MATCNN INSTANTVISION