Roof-edge preserving image smoothing based on MRFs

In this paper, we present a technique for accelerating the bilateral filtering using GPGPU. Bilateral filtering is a tool for an image smoothing with edge preserving properties. It serves as a mixture of domain and range filters. Domain filter suppresses Gaussian noise while range filter maintains sharp edges. Bilateral filtering is a nonlinear filtering in which the filter kernel must be computed pixel by pixel. Therefore conventional fast Fourier transform technique cannot be used to accelerate the bilateral filtering. Instead, general purpose GPU is used as a parallel machine to reduce time consuming of the bilateral filtering. We will show the experimental results by comparing the computation time of CPU and GPU. It was cleared that, from the experimental results, GPU outperformed the CPU in terms of computation time.

Download Full-text

Edge Preserving Range Image Smoothing by Rotated Bilateral Sampling

2015 4th Eastern European Regional Conference on the Engineering of Computer Based Systems ◽

10.1109/ecbs-eerc.2015.22 ◽

2015 ◽

Cited By ~ 1

Author(s):

Viktor Kovacs ◽

Kristof Csorba ◽

Gabor Tevesz

Keyword(s):

Range Image ◽

Image Smoothing ◽

Edge Preserving

Download Full-text

The effect of edge-preserving image smoothing on automatic colonic polyp detection for CT colonography

10.1117/12.653985 ◽

2006 ◽

Cited By ~ 2

Author(s):

David Pilkinton ◽

Ingmar Bitter ◽

Ronald M. Summers ◽

Shannon Campbell ◽

J. R. Choi ◽

...

Keyword(s):

Ct Colonography ◽

Colonic Polyp ◽

Image Smoothing ◽

Polyp Detection ◽

Edge Preserving ◽

Colonic Polyp Detection

Download Full-text

A Generalized Framework for Edge-Preserving and Structure-Preserving Image Smoothing

Proceedings of the AAAI Conference on Artificial Intelligence ◽

10.1609/aaai.v34i07.6830 ◽

2020 ◽

Vol 34 (07) ◽

pp. 11620-11628

Author(s):

Wei Liu ◽

Pingping Zhang ◽

Yinjie Lei ◽

Xiaolin Huang ◽

Jie Yang ◽

...

Keyword(s):

Penalty Function ◽

State Of The Art ◽

Superior Performance ◽

Image Smoothing ◽

Edge Preserving ◽

Structure Preserving ◽

Smoothing Operator ◽

Optimization Framework ◽

Generalized Framework ◽

Smooth Optimization

Image smoothing is a fundamental procedure in applications of both computer vision and graphics. The required smoothing properties can be different or even contradictive among different tasks. Nevertheless, the inherent smoothing nature of one smoothing operator is usually fixed and thus cannot meet the various requirements of different applications. In this paper, a non-convex non-smooth optimization framework is proposed to achieve diverse smoothing natures where even contradictive smoothing behaviors can be achieved. To this end, we first introduce the truncated Huber penalty function which has seldom been used in image smoothing. A robust framework is then proposed. When combined with the strong flexibility of the truncated Huber penalty function, our framework is capable of a range of applications and can outperform the state-of-the-art approaches in several tasks. In addition, an efficient numerical solution is provided and its convergence is theoretically guaranteed even the optimization framework is non-convex and non-smooth. The effectiveness and superior performance of our approach are validated through comprehensive experimental results in a range of applications.

Download Full-text