A Fast and Accurate Prediction of BPG Compression Parameters in Optimal Operation Point Neighbourhood for Three-channel Noisy Images | Chapter 7 | Fundamental Research and Application of Physical Science Vol. 8

This stage considers the lossy compression of three-channel images debased by additive silvery Gaussian noise. The better portable drawings (BPG) encoder that has three different movement modes is studied. As for additional encoders, a noise draining effect is observed and it influences the encoder performance. For certain environments (image complicatedness and noise intensity), the supposed optimal movement point (OOP) may exist. The main feature of OOP is that the quality of a compacted image (calculated concerning the true countenance) is better than the quality of the corresponding strident image in accordance with a given metric (either common such as mean square mistake or some visual value metric). Knowledge having to do with the OOP existence and compression limits in it for a given strident image can be valuable. In the case of OOP existence, it wash to compress a given representation in OOP or its community. If OOP does not exist, a more “careful” lossy compression is expedient. The question is that in practice the valid image is not available, and, so, it is impossible to reckon full-reference versification for compressed and roar-free images. Meanwhile, we show in this paper that it is likely to predict condensation parameters for a given cacophonous image earlier. Moreover, this prediction is fast and accurate enough for a conclusion undertaking on which coder parameter to set to reach OOP. In addition, in this place paper, we demonstrate that it is likely to predict the compression limits not only in OOP but also in allure neighborhood. This allows for endeavor more reliable conclusions. Prediction opportunities are shown for two optic quality versification and three typical modes of the BPG coder movement: 4:4:4, 4:2:2, and 4:2:0. One novel way of prognosis accuracy improvement established the joint use of two input limits is shown.

Author(s) Details:

Vladimir Lukin,
National Aerospace University, Kharkov, Ukraine.

Bogdan Kovalenko,
National Aerospace University, Kharkov, Ukraine.

Benoit Vozel,
University of Rennes, Lannion, France.

Please see the link here: https://stm.bookpi.org/FRAPS-V8/article/view/11676