Level Crossing Statistics: A Simple Image Analysis Approach to Formation and Print Quality Assessment, 1996 Process & Product Quality Conference Proceedings

R.J. Trepanier [1] , B.D. Jordan [2] , N.G. Nguyen [2]

[I] OpTest Equipment Inc.

[2] Pulp and Paper Research Institute of Canada

Sheet formation and solid print mottle are described in terms of textured images. The traditional method to quantify the non-uniformity of a textured image is to determine the standard deviation of the texture variations. However, first order statistics, such as the standard deviation, do not account for the size distribution of the texture non-uniformities. This paper presents a simple technique employing level crossing statistics to quantify the graininess of the textures in print mottle or formation. Specifically the length of the median-density contour line per unit area of the image is calculated The result is called the Specific Perimeter. The Specific Perimeter has been previously derived and mathematically related to the power spectrum and the autocorrelation function. This paper discusses the Specific Perimeter and considers its application.