论文信息 - TRI-DCT BASED FAST BACK PROPAGATION ALGORITHM

TRI-DCT BASED FAST BACK PROPAGATION ALGORITHM

 Abstract: In Artificial neural networks the number of neurons in hidden layers are fewer than input and output layer consequently ANN is used for image compression and decompression. Among the ANN's back propagation neural network algorithm (BP) is finest for image compression but training time is long. So the proposed DCT based fast back propagation neural network reduces the training time and improve the compression ratio by changing the input layer and hidden layer neurons. Time of conversation is further reduced by choosing a best learning rate parameter η. Finally DCT based fast back propagation neural network results such as compression ratio (CR) and peak signal to noise ratio (PSNR) are computed and compared with BP results.

K. Chiranjeevi | G. Visalakshi

[1] Fatima B. Ibrahim. Image Compression using Multilayer Feed Forward Artificial Neural Network and DCT , 2010 .

[2] Omaima N. A. AL-Allaf,et al. Improving the Performance of Backpropagation Neural Network Algorithm for Image Compression/Decompression System , 2010 .

[3] Tai-Hoon Cho,et al. Fast backpropagation learning using steep activation functions and automatic weight reinitialization , 1991, Conference Proceedings 1991 IEEE International Conference on Systems, Man, and Cybernetics.

[4] J. Todd. Book Review: Digital image processing (second edition). By R. C. Gonzalez and P. Wintz, Addison-Wesley, 1987. 503 pp. Price: £29.95. (ISBN 0-201-11026-1) , 1988 .

[5] Liu Yang,et al. An Image Compressing Algorithm Based on Classified Blocks with BP Neural Networks , 2008, 2008 International Conference on Computer Science and Software Engineering.

[6] Richard P. Lippmann,et al. An introduction to computing with neural nets , 1987 .

[7] B. Yegnanarayana,et al. Artificial Neural Networks , 2004 .

[8] N. K. Ibrahim,et al. Artificial neural network approach in radar target classification , 2009 .

[9] Anastasios N. Venetsanopoulos,et al. Artificial neural networks - learning algorithms, performance evaluation, and applications , 1992, The Kluwer international series in engineering and computer science.