Selective image sharpening by simultaneous nonlinear-diffusion process with spatially varying parameter presetting

Authors:
Takahiro Saito;Shigemitsu Anyoji;Takashi Komatsu
Affiliations:
Department of Electrical, Electronics and Information Engineering, High-Tech Research Center, Kanagawa University, Yokohama, Japan;Department of Electrical, Electronics and Information Engineering, High-Tech Research Center, Kanagawa University, Yokohama, Japan;Department of Electrical, Electronics and Information Engineering, High-Tech Research Center, Kanagawa University, Yokohama, Japan
Venue:
PCM'04 Proceedings of the 5th Pacific Rim Conference on Advances in Multimedia Information Processing - Volume Part II
Year:
2004

Citing 7
Cited 0

Precision Edge Contrast and Orientation Estimation

IEEE Transactions on Pattern Analysis and Machine Intelligence
Scale-Space and Edge Detection Using Anisotropic Diffusion

IEEE Transactions on Pattern Analysis and Machine Intelligence
Feature-oriented image enhancement using shock filters

SIAM Journal on Numerical Analysis
Image selective smoothing and edge detection by nonlinear diffusion

SIAM Journal on Numerical Analysis
A Variational Method in Image Recovery

SIAM Journal on Numerical Analysis
Digital Picture Processing

Digital Picture Processing
Forward-and-backward diffusion processes for adaptive image enhancement and denoising

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Previously we have presented a PDE-based method for selective image sharpening. Our method works as the simultaneous nonlinear-diffusion process composed of a nonlinear-diffusion term, a fidelity term and a peaking term, and it sharpens blurred edges while smoothing out noisy variations. However, our method has the problem that it does not satisfactorily sharpen complex image-structures such as T-shaped edges and textures. This paper copes with the problem, and improves selective image sharpening. As the preprocess of our PDE-based method, this paper introduces a step to classify each pixel into two categories on the basis of mid-scale image-features contained in the image-gradient field. The classification results are then utilized to preset the parameters characterizing our PDE-based method spatially varyingly.