Video Decompression Technology in Information and Communication Technologies
Електронного архіву Харківського національного університету радіоелектроніки (Open Access Repository of KHNURE)
Переглянути архів Інформація| Поле | Співвідношення | |
| Title |
Video Decompression Technology in Information and Communication Technologies
|
|
| Creator |
Ryabukha, Yu.
Krivonos, V. Hahanova, A. V. |
|
| Subject |
low-frequency DC component
Recovery of vector of significant subbands |
|
| Description |
It is proposed to introduce new methods of video data compression for more efficient use of wireless technology. Therefore, we propose the method of reconstructing digital static images based on the restoration of transforms. There is a technology of renovating values of vector of significant subbands of the nonuniform DCT spectrum on a known code and base; the vector of scaling components based on the decoding of the first zero series and Bodo codes; lowfrequency DC component using a statistical code. The decompression method proposed allows restoring an image without making the information loss for a given confidence level. An intensive growth of video data, which significantly advance the production capacity of the existing wireless telecommunication systems, generates a need in their compact representation. Existing methods of video data compression, according to the characteristics of the compression ratio, allow transmitting images in real time, however, with a significant loss of quality. One of the areas that can improve the performance of existing wireless technologies in transferring and processing video data is based on the introduction of new methods of compression, which, in turn, will reduce the cost of technical modernization of telecommunication systems. Hence, the development of video coding method to reduce their volume in telecommunication systems is an actual scientific and application task. Therefore, we propose method of compressing video data with controlled quality loss based on JPEG platform [1, 2]. This allows to reduce the amount of information at the desired level of reliability of the image. Image processing starts with changing the RGB color model to YCrCb model, with further segmentation, i.e. partitioning the image into blocks of 8 × 8 pixels. Further, a discrete cosine transform is applied to each block (DCT). Then, quantization and non-uniform decrease of DCT coefficient values are applied to the DCT transform. The next step is the formation of dimensional vector of n Y component from the transforms using diagonal scanning. A lowfrequency DC component is allocated individually from the one-dimensional vector formed, and the vector of significant n 1 Y − subbands and the vector of scaling 1 nG − components are formed from the remaining components. Vector of significant subbands is a sequence of non-uniform values of DCT frequency spectrum. Components of the vector n 1 G − indicate the length of the corresponding sub-band. This method allows gaining an advantage over the JPEG standard, regarding the processing of medium and highly saturated images. In addition, there is a need in reconstructing an image, which won’t exceed the complexity of coding and will restore the image without making mistakes. Hence, the objective of research is to develop a method for the reconstruction of compressed video data using the pre-transformation.
|
|
| Date |
2016-09-06T09:19:03Z
2016-09-06T09:19:03Z 2014 |
|
| Type |
Article
|
|
| Identifier |
Ryabukha Yu. Video Decompression Technology in Information and Communication Technologies /Yu. Ryabukha, Vladimir Krivonos, Anna Hahanova //Proceedings of IEEE East-West Design & Test Symposium (EWDTS’2014)
http://openarchive.nure.ua/handle/document/2137 |
|
| Language |
en
|
|
| Publisher |
EWDTS
|
|