Neural net based digital halftoning of images

Various novel techniques for digital image halftoning are presented, performing nonstandard quantization subject to a fidelity criterion. Hopfield-type networks can be used for this task, minimizing a frequency-weighted mean squared error between the input (continuous-tone) and the output (bilevel)...

Full description

Saved in:
Bibliographic Details
Published in1988 IEEE International Symposium on Circuits and Systems (ISCAS) pp. 507 - 510 vol.1
Main Author Anastassiou, D.
Format Conference Proceeding
LanguageEnglish
Published IEEE 1988
Subjects
Artificial neural networks
Differential equations
Digital images
Displays
Frequency
Magnetic analysis
Neural networks
Neurons
Nonlinear dynamical systems
Quantization
Online AccessGet full text

Cover

More Information
Summary:Various novel techniques for digital image halftoning are presented, performing nonstandard quantization subject to a fidelity criterion. Hopfield-type networks can be used for this task, minimizing a frequency-weighted mean squared error between the input (continuous-tone) and the output (bilevel) image. A novel kind of massively parallel analog network (the differential neural network) is introduced and shown to be appropriate for this task. This kind of network contains a nonmonotonic nonlinearity in lieu of the sigmoid function.< >
DOI:10.1109/ISCAS.1988.14975