|
||||||||||
|
DOI: 10.17587/it.25.475-481 S. Sh. Fahmi, Professor, e-mail: shakeebf@mail.ru, St. Petersburg state electrotechnical University V. I. Ulyanov (Lenin) "LETI", Solomenko Institute of Transport Problems of the Russian academy of sciences, N. G. Ryjov, Assistant Professor, e-mail: ngryzhov@etu.ru, St. Petersburg state electrotechnical University V. I. Ulyanov (Lenin) "LETI", Y. A. Hasan, post-graduate student, e-mail: midocom@mail.ru, St. Petersburg state electrotechnical University. V. I. Ulyanov (Lenin) "LETI", E. S. Kalinina, Assistant Professor, e-mail: elenakali21@yandex.ru, Saint-Petersburg University of state fire service of Emercom of Russia It is well known that the image processing standards (JPEG) and video processing standards (MPEG), based on the use of spectral cosine transformation, do not meet the growing requirements for high quality video transmission obtained from surveillance cameras in transport. The reason for this is the large amount of visual data and the fact that the optimization of the video codec should be done (at least) on three parameters: bit rate, distortion and computational complexity. Therefore, the task arises to develop new image compression algorithms that are best aligned with existing standards. One of the algorithms for improving the process of coding and decoding is considered in this paper. It differs from the existing ones in that the sizes of blocks subjected to cosine transformation can be arbitrary and are selected taking into account the specifics of transport subjects. P. 475–481 |