Title :
A Wavelet Neural Network for Edge Detection
Abstract :
Wavelet neural networks have the advantages of good localization characteristics and adaptive learning virtue. A wavelet neural network, in this paper, is used for edge detection in digital images. Image features are extracted as the input of the network from local statistics of the image. The experimental results indicate that the proposed method is able to effectively complete the task of edge detection.
Keywords :
Adaptive signal processing; Artificial neural networks; Digital images; Feature extraction; Gray-scale; Image edge detection; Neural networks; Pixel; Statistics; Wavelet transforms; difference features; edge detection; median features; wavelet neural network;
Conference_Titel :
Image and Signal Processing, 2008. CISP '08. Congress on
Conference_Location :
Sanya, China
Print_ISBN :
978-0-7695-3119-9
DOI :
10.1109/CISP.2008.110
