Presentation 2012-12-13
Image approximation with root images of morphological filters and its application to image recovery
Makoto NAKASHIZUKA, Yu ASHIHARA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Morphology filter / signal approximation / image recovery / image denoising
Keyword(in Japanese) (See Japanese page)
Keyword(in English) This report proposes an image approximation method with root signals of morphological filters and its application to image recovery. In morphological image processing, an image is transformed into a subset of the three-dimensional space, of which axises are intensity and two-dimensional coordinates. Opening and closing, which are typical morphological filters, obtains an approximation of an image as a union of translated structuring elements in the three dimensional space. Opened or closed images are root signals of opening or closing. However, the opening and closing do not approximate images in terms of minimum squared error. In this report, we propose an approximation method with the root signals of the morphological filter by using soft morphological filters and framework of regularization. In experiments, we apply the propose method to denoising of noisy images corrupted by Gaussian noises and compare the denoising results with results obtained by regularization method that imposes the penalty on the intensity differences among image pixels
Paper # SIS2012-39
Date of Issue

Conference Information
Committee SIS
Conference Date 2012/12/6(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Smart Info-Media Systems (SIS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Image approximation with root images of morphological filters and its application to image recovery
Sub Title (in English)
Keyword(1) This report proposes an image approximation method with root signals of morphological filters and its application to image recovery. In morphological image processing, an image is transformed into a subset of the three-dimensional space, of which axises are intensity and two-dimensional coordinates. Opening and closing, which are typical morphological filters, obtains an approximation of an image as a union of translated structuring elements in the three dimensional space. Opened or closed images are root signals of opening or closing. However, the opening and closing do not approximate images in terms of minimum squared error. In this report, we propose an approximation method with the root signals of the morphological filter by using soft morphological filters and framework of regularization. In experiments, we apply the propose method to denoising of noisy images corrupted by Gaussian noises and compare the denoising results with results obtained by regularization method that imposes the penalty on the intensity differences among image pixels
1st Author's Name Makoto NAKASHIZUKA
1st Author's Affiliation Faculty of Engineering Chiba Institute of Technology()
2nd Author's Name Yu ASHIHARA
2nd Author's Affiliation Graduate School of Engineering Science Osaka University
Date 2012-12-13
Paper # SIS2012-39
Volume (vol) vol.112
Number (no) 348
Page pp.pp.-
#Pages 6
Date of Issue