A CRITERION FOR THE QUANTIZATION OF LINEDRAWING DATA.

Abstract

The digital transmission of line-drawing data requires that the data first be quantized. Since quantization tends to degrade the original data, it is desirable to employ as fine a quantization as possible. A criterion is developed for enabling one to determine the coarsest possible quantization that will preserve the original data to within a specified tolerance. In the criterion developed, line drawings are thought of as thin elastic beams, which can be characterized by means of their strain-energy density functions. Quantization is seen to result in a spreading of these density functions, creating an uncertainty region in an energy space, the uncertainty decreasing with finer quantization. Of the possible continuous reconstructions that can be generated from the quantized data, there is one that is of particular interest - the minimum-energy curve. An appropriate quantization is determined by requiring that the total energies of the minimum-energy curve and the original curve be within a specified tolerance.

Document Details

Document Type
Technical Report
Publication Date
May 01, 1965
Accession Number
AD0621086

Entities

People

  • Jeremy M. Glass

Organizations

  • New York University

Tags

Communities of Interest

  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Uncertainty

Fields of Study

  • Physics

Readers

  • Approximation Theory.
  • Computer Vision.

Technology Areas

  • Space