Discriminating color space selection for edge detection using multiscale product wavelet transform
Edge detection is an important low level image processing step, which can influence the final results. In this paper, an effective method, named the multiscale product wavelet transform, is proposed for edge detection. Although the one scale wavelet transform is a universal method, it is not suitable for real complex images because this method cannot efficiently detect edges. Therefore, the output edge map suffers from a considerable amount of false and double edges. An extension to the one scale wavelet transform approach is to use multiscale transform and perform the product of these scales. The use of the resulting product improves the localization of detected edges and dilutes noise. Unlike others using the multiscale transform, the proposed method applies the multiscale product wavelet transform on the most discriminating color space. This has given the best results in term of edges using a robust statistical analysis. The performance of the method is tested on a database containing 500 real images, which present very complex information.
Keywords: Edge detection; Discriminating color space; Dyadic wavelet transform; Multiscale wavelet product; ROC analysis.
Download Full-Text
ABOUT THE AUTHORS
Sihem Charaâ
Sihem Charaâ is a PhD student in National Engineering High School of Tunis (ENIT). She earned her Bachelor degree in computer science from Sciences High University of Tunis, Tunisia, and her Master degree in Signal and image processing from National Engineering High School of Tunis. Her PhD is specializing in Image Processing.
Noureddine Ellouze
Pr. Noureddine Ellouze received a PhD degree in 1977 from l’Institut National Polytechnique at Paul Sabatier University (Toulouse, France), and Electronic Engineer Diploma from ENSEEIHT in 1968 at the same university. He have served as Director of Electrical Department at ENIT from 1978 to 1983, General Manager and President of the Research Institute on Informatics and Telecommunications (IRSIT) from 1987 to 1990, President of the same Institute from 1990 to 1994. He is now Director of Signal Processing Research Laboratory (LSTS) at ENIT and is in charge of Control and Signal Processing Master degree at ENIT. Pr. Ellouze is IEEE fellow since 1987, he directed multiple Master thesis and PhD thesis and published over 200 scientific papers in journals and conference proceedings. He is chief editor of the scientific journal Annales Maghrébines de l’Ingénieur. His research interests include Neural Networks and Fuzzy Classification, Pattern Recognition, Signal Processing and Image Processing applied in Biomedical, Multimedia, and Man Machine Communication.
Sihem Charaâ
Sihem Charaâ is a PhD student in National Engineering High School of Tunis (ENIT). She earned her Bachelor degree in computer science from Sciences High University of Tunis, Tunisia, and her Master degree in Signal and image processing from National Engineering High School of Tunis. Her PhD is specializing in Image Processing.
Noureddine Ellouze
Pr. Noureddine Ellouze received a PhD degree in 1977 from l’Institut National Polytechnique at Paul Sabatier University (Toulouse, France), and Electronic Engineer Diploma from ENSEEIHT in 1968 at the same university. He have served as Director of Electrical Department at ENIT from 1978 to 1983, General Manager and President of the Research Institute on Informatics and Telecommunications (IRSIT) from 1987 to 1990, President of the same Institute from 1990 to 1994. He is now Director of Signal Processing Research Laboratory (LSTS) at ENIT and is in charge of Control and Signal Processing Master degree at ENIT. Pr. Ellouze is IEEE fellow since 1987, he directed multiple Master thesis and PhD thesis and published over 200 scientific papers in journals and conference proceedings. He is chief editor of the scientific journal Annales Maghrébines de l’Ingénieur. His research interests include Neural Networks and Fuzzy Classification, Pattern Recognition, Signal Processing and Image Processing applied in Biomedical, Multimedia, and Man Machine Communication.