DocumentCode
1803322
Title
Neural net based digital halftoning of images
Author
Anastassiou, Dimitris
Author_Institution
Dept. of Electr. Eng., Columbia Univ., New York, NY, USA
fYear
1988
fDate
7-9 Jun 1988
Firstpage
507
Abstract
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
Keywords
neural nets; picture processing; Hopfield-type networks; differential neural network; digital halftoning; fidelity criterion; frequency-weighted mean squared error; images; massively parallel analog network; nonmonotonic nonlinearity; nonstandard quantization; Artificial neural networks; Differential equations; Digital images; Displays; Frequency; Magnetic analysis; Neural networks; Neurons; Nonlinear dynamical systems; Quantization;
fLanguage
English
Publisher
ieee
Conference_Titel
Circuits and Systems, 1988., IEEE International Symposium on
Conference_Location
Espoo
Type
conf
DOI
10.1109/ISCAS.1988.14975
Filename
14975
Link To Document