Skip navigation

Analysis of Motion Blur Using Double Discrete Wavelet Transform

Event Date: April 5, 2012
Speaker: Dr. Keigo Hirakawa
Speaker Affiliation: University of Dayton
Sponsor: Communications, Networking, Signal & Image Processing
Time: 1:30 PM
Location: EE 118
Contact Name: Professor Jan Allebach
Contact Phone: (765) 494-3535
Contact Email:

Object motion causes spatially varying blur. Estimating such a type of blur from a single image is an ill-posed problem that is difficult to solve. In this paper, we introduce the notion of double discrete wavelet transform (DDWT) designed to sparsify the blurred image and blur kernel simultaneously. Based on DDWT analysis, we are able to accurately estimate motion blur kernels and recover the latent sharp image. The blind image deblurring solution we propose handles spatially varying motion blurs effectively and efficiently.

Keigo Hirakawa joined University of Dayton as Assistant Professor of Electrical and Computer Engineering in 2010. Prior to UD, he was with Harvard University as a Research Associate of Department of Statistics and School of Engineering and Applied Sciences. He simultaneously earned Ph.D. in Electrical and Computer Engineering from Cornell University and M.M. in Jazz Performance from New England Conservatory of Music. He received his B.S. in Electrical Engineering from Princeton University. Prof. Hirakawa has published in the literature of engineering, computer science, and statistics. His research focuses on algorithmic development of image processing, computer vision, biomedical imaging, and sensor designs. He is best known for his expertise in digital camera designs, and his contributions span color science, estimation theory, statistical modeling, and wavelet theory. He has received a number of recognitions, including a paper award from IEEE ICIP 2007 and keynote presentations at IS&T CGIV, PCSJ-IMPS, and CSAJ, and IAPR CCIW. He is an associate editor for SPIE/IS&T Journal of Electronic Imaging, and serves on the organization committees of IEEE ICIP 2012.