Design of FIR Bilevel Laplacian-of-Gaussian filter

Authors:
Soo-Chang Pei;Ji-Hwei Horng
Affiliations:
Department of Electrical Engineering, National Taiwan University, Taipei, Taiwan, ROC;Department of Information Management, National Kaohsiung University of Applied Sciences at Kinmen, Taiwan, ROC
Venue:
Signal Processing
Year:
2002

Citing 5
Cited 1

Detection of Intensity Changes with Subpixel Accuracy Using Laplacian-Gaussian Masks

IEEE Transactions on Pattern Analysis and Machine Intelligence
Fast Convolution with Laplacian-of-Gaussian Masks

IEEE Transactions on Pattern Analysis and Machine Intelligence
The Laplacian-of-Gaussian kernel: a formal analysis and design procedure for fast, accurate convolution and full-frame output

Computer Vision, Graphics, and Image Processing
Refining edges detected by a LoG operator

Computer Vision, Graphics, and Image Processing
Multidimensional Digital Signal Processing

Multidimensional Digital Signal Processing

A fast multiplier-less edge detection accelerator for FPGAs

Proceedings of the 2010 ACM Symposium on Applied Computing

Quantified Score

Hi-index	0.08

Visualization

Abstract

The Laplacian-of-Gaussian (LOG) filter is an optimal edge detector, but it is computationally inefficient. In this paper, we propose the bilevel Laplacian-of-Gaussian (BLoG) filter to approximate the LoG filter. This approximation is formulated as an optimization problem and solved by the gradient descent algorithm. Only two multiplications per pixel are required in convolving with the BLoG filter. The computation is fast and it produces satisfactory results. The applications of ID BLoG filter to corner detection and 2D BLoG filter to edge detection are also presented.