Browse Prior Art Database

Image Contrast Enhancement by Local Maximum and Local Minimum Operators

IP.com Disclosure Number: IPCOM000049061D
Original Publication Date: 1982-May-01
Included in the Prior Art Database: 2005-Feb-09
Document File: 3 page(s) / 35K

Publishing Venue

IBM

Related People

Jackson, PH: AUTHOR

Abstract

This article describes s method for enhancing line structure within images by increasing its contrast relative to the background.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 55% of the total text.

Page 1 of 3

Image Contrast Enhancement by Local Maximum and Local Minimum Operators

This article describes s method for enhancing line structure within images by increasing its contrast relative to the background.

The use of local maximum and local minimum operations in removing "salt and pepper" noise in digital images has been described by Nakagawa and Rosenfeld (1). These operators are grey scale generalizations of binary image shrink/expand operators. By applying first local min and then local max operators to an image, a new image is produced in which every peak (of less than 2 pixels (picture elements) width) has been "flattened", i.e., had its value reduced to the level of the surrounding pixels. By applying first local max and then local min, troughs have their minima brought up to the surrounding level.

The width of the peaks and troughs which are reduced is determined by the number of local max and local min operations performed. Thus for peaks, image I is operated on first by n local min operations and then by n local max operations, producing image P, vis: P equals LMAX/n/ LMIN/n/ I and for troughs, producing image T, T equals LMIN/n/ LMAX/n/ I

These operations suppress peaks and troughs of width less than or equal to 2n.

Following this work, Pass (2) noted that these peaks and troughs may not be just noise, and could be of interest. The peaks are given by the subtraction of P from I, and the troughs are given by the subtraction of T from I. These were used for scene analysis.

The application of this to contrast enhancement is as follows.

If the peaks are added back in to the original image, and the troughs are subtracted from the original image, the result is...