Title :
Neural concurrent subsampling and interpolation for images
Author :
Kim, Jong-Ok ; Choi, Byung-Tae ; Morales, Aldo ; Ko, Sung-Jea
Author_Institution :
Dept. of Electron. Eng., Korea Univ., Seoul, South Korea
Abstract :
This paper presents a new method for image subsampling and interpolation based on the feedforward neural network (FNN). The proposed technique employs a single FNN with three hidden layers for both subsampling and interpolation, providing the advantages of high speed, parallel processing capability, and good image reproduction quality. Experimental results show that the proposed technique exhibits an increased performance over conventional ones
Keywords :
feedforward neural nets; image sampling; interpolation; parallel processing; experimental results; feedforward neural network; hidden layers; high speed method; image interpolation; image reproduction quality; image subsampling; neural concurrent subsampling; parallel processing; performance; Educational institutions; Feedforward neural networks; Image coding; Image quality; Image reconstruction; Image restoration; Interpolation; Neural networks; Parallel processing; Spatial resolution;
Conference_Titel :
TENCON 99. Proceedings of the IEEE Region 10 Conference
Conference_Location :
Cheju Island
Print_ISBN :
0-7803-5739-6
DOI :
10.1109/TENCON.1999.818674
