An overview of LULU operators and discrete pulse transform for image analysis

Abstract

This paper presents an overview of LULU operators and discrete pulse transform (DPT). Data extraction from signals and images is a popular area of research. Different methods are being used for data extraction such as different types of linear and nonlinear operators. LULU operator is one of the most important rank selector nonlinear filters used for data analysis which is widely being used for signal analysis, especially in impulse noise filtering. It is computationally more efficient and the behaviour of the operator is simple to describe. Based on the composition of different orders of LULU operators, DPT on multi-resolution is defined, which describes the sequences into pulses with different magnitudes. DPT allows a multi-resolution measure of roughness of images and sequences. It is a powerful technique for image analysis and can also be used for the estimation of standard deviation of a random distribution.

Keywords:

• discrete pulse transform
• impulse noise
• object extraction

This work is supported by E-Science grant funded by the Ministry of Science, Technology and Innovation, Government of Malaysia (no. 01-02-02-SF0064). We would like to thank Professor Roumen Anguelov and Ms Inger Fabris-Rotelli for their support for providing us with information on 2D DPT.