Deep Learning Based EBCOT Source Symbol Prediction Technique for JPEG2000 Image Compression Architecture
Journal
2020 IEEE International Conference on Visual Communications and Image Processing, VCIP 2020
Pages
226-229
Date Issued
2020
Author(s)
Abstract
In this work, an efficient and robust learning-based JPEG2000 architecture is proposed. It uses machine learning techniques for predicting and encoding the decision bit in the embedded block coding with optimized truncation (EBCOT) process. First, we apply non-locally weighted ridge regression to predict the quantized wavelet coefficients in the LL subband. Then, during the EBCOT process, we perform inter/intra subband prediction and inter/intra bit plane symbol prediction to estimate the activity of the decision bit using the deep learning architecture. Then, the binary prediction result is treated as an additional context and the decision bit is eventually coded using an advanced context-based adaptive binary arithmetic coder. Simulations show that the proposed framework provides the same visual quality as conventional codecs with as much as 30% bitrate savings. ? 2020 IEEE.
Subjects
Binary codes; Digital image storage; Forecasting; Image compression; Learning systems; Regression analysis; Visual communication; Binary arithmetic coders; Embedded block coding with optimized truncation; JPEG2000 image compression; Learning architectures; Machine learning techniques; Prediction techniques; Ridge regression; Wavelet coefficients; Deep learning
Type
conference paper