Title :
An error diffusion neural network for digital image halftoning
Author :
Shoop, Barry L. ; Ressler, Eugene K.
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., US Mil. Acad., West Point, NY, USA
fDate :
31 Aug-2 Sep 1995
Abstract :
A novel technique for digital image halftoning is proposed, based on a symmetric error diffusion algorithm and a new form of artificial neural network. Using an error diffusion neural network, all pixel quantization decisions are computed in parallel and therefore visual artifacts resulting from the causality of the diffusion filter in classical error diffusion techniques are reduced and the resulting halftoned image quality is improved
Keywords :
Hopfield neural nets; computer vision; filtering theory; image processing; quantisation (signal); sensor fusion; Hopfield type neural network; diffusion filter; digital image halftoning; error diffusion neural network; parallel processing; pixel quantization; symmetric error diffusion; Computer errors; Digital images; Distortion measurement; Filtering; Hopfield neural networks; Low pass filters; Military computing; Neural networks; Pixel; Quantization;
Conference_Titel :
Neural Networks for Signal Processing [1995] V. Proceedings of the 1995 IEEE Workshop
Conference_Location :
Cambridge, MA
Print_ISBN :
0-7803-2739-X
DOI :
10.1109/NNSP.1995.514917
