The following information was submitted:
Transactions: INTERNATIONAL JOURNAL of APPLIED MATHEMATICS AND INFORMATICS
Transactions ID Number: 20-423
Full Name: Peter Ndajah
Position: Student
Age: ON
Sex: Male
Address: Graduate School of Science and Technology, Niigata University, 8050, Ikarashi 2-no-cho, Nishi-ku, Niigata, 950-2181
Country: JAPAN
Tel: +818031495685
Tel prefix:
Fax:
E-mail address: ndajah@telecom0.eng.niigata-u.ac.jp
Other E-mails: ndajah@gmail.com
Title of the Paper: scaled image edge detection based on the total variation functional
Authors as they appear in the Paper: Peter Ndajah, Hisakazu Kikuchi
Email addresses of all the authors: ndajah@telecom0.eng.niigata-u.ac.jp, kikuchi@eng.niigata-u.ac.jp
Number of paper pages: 10
Abstract: We present total variation anisotropic edge detection. We derive the total variation functional from measure theory, distribution theory and vector gradient method. The Euler-Lagrange equation of the total variation functional gives a steady state equation. The steady state equation acts as an anisotropic filter on an image. The total variation filtered images are compared to Laplacian filtered images. A subsequent application of the zero crossing algorithm works quite well for the traditional Marr-Hildreth method but gives poorer results for the total variation filtered images. It was found that thresholding methods work better and saves computational time. Also, our results show that total variation edge detection overcomes some drawbacks associated with the Marr-Hildreth method.
Keywords: Anisotropic, Euler-Lagrange, LoG, Total Variation, Laplace
EXTENSION of the file: .ps
Special (Invited) Session: Total Variation Image Edge Detection
Organizer of the Session: 650-438
How Did you learn about congress:
IP ADDRESS: 133.35.135.65
