Title :
JPEG post processing image enhancement
Author :
Lopez, Alex ; Torres, Angel ; Hunt, Shawn
Author_Institution :
Eastman Kodak Co., Rochester, NY, USA
Abstract :
A nonlinear filter based on the feedforward neural network topology is presented. This study was undertaken to investigate the usefulness of `smart´ filters in image post processing. The filter has shown to be useful in recovering high frequencies, such as those lost during the JPEG compression-decompression process. The filtered images have a higher signal to noise ratio, and a higher perceived image quality. Simulation studies showing examples of the high frequency recovery, and the statistical properties of the filter are given
Keywords :
data compression; feedforward neural nets; filtering theory; image coding; image enhancement; image restoration; nonlinear filters; JPEG compression-decompression process; JPEG post processing image enhancement; feedforward neural network topology; filtered images; high frequency recovery; image quality; nonlinear filter; signal to noise ratio; smart filters; statistical properties; Feedforward neural networks; Frequency; Image coding; Image enhancement; Image quality; Network topology; Neural networks; Nonlinear filters; Signal to noise ratio; Transform coding;
Conference_Titel :
Neural Networks,1997., International Conference on
Conference_Location :
Houston, TX
Print_ISBN :
0-7803-4122-8
DOI :
10.1109/ICNN.1997.614018
