Browse Prior Art Database

Histogram Techniques for Extraction of Information From Greyscale Images

IP.com Disclosure Number: IPCOM000039771D
Original Publication Date: 1987-Aug-01
Included in the Prior Art Database: 2005-Feb-01

Publishing Venue

IBM

Related People

Authors:
Trisoglio, AR [+details]

Abstract

Techniques are described whereby histograms, used for extraction of information from greyscale images, are improved through the use of filtering, smoothing, detection of minima, comparing and thresholding concepts. Also discussed are methods of improving signal-to-noise ratios, edge detection and recombining concepts. The techniques denote an improvement over prior approaches, such as the use of spatial techniques or correlation methods. The technique expands the use of histograms, so as to cover larger ranges of general information extraction. Typically, the histogram of a greyscale image is constructed by finding the total number of pixels at each greylevel and has several local maxima and local minima.