Font Size: a A A

The Research Of JPEG2000 Encoder Based On DSP

Posted on:2009-02-24Degree:MasterType:Thesis
Country:ChinaCandidate:Q XueFull Text:PDF
GTID:2308330464470361Subject:Communication engineering
Abstract/Summary:PDF Full Text Request
With the development of multimedia technology and network, people demand higher in the digital image quality, size and its application, and hope to transfer and storage substantial images with the limited space and bandwidth resources. Meanwhile, the reconstructed image of different quality or resolution is also been wanted to be obtained according to the actual needs. Replacing discrete cosine transform with wavelet transform which is a great mend of JPEG2000 can guarantee the image not only in good quality, forgiveness in the transmitting and also in supporting coding region of interest. Therefore, it is quite necessary to study the arithmetic of JPEG2000.After studying the standard of the JPEG2000, this thesis researched mainly the approaches to realizing wavelet transform and arithmetic coding on the DSP hardware platform. Following are the main contents of those researches.(1) After study on standards of JPEG2000 and its validation, this thesis discussed in detail about the wavelet transform and arithmetic entropy coding algorithm to the core of the JPEG2000 codec system and its workflow. Programmed the core code algorithm based on C++ language, and has been realized on PC machines.(2) Having studied the hardware characteristics of DSP6000, the article took TI’s TMS320DM642 DSK as an embedded platform, transplanted the code algorithm to the to make full use of on-chip memory to choose, adjust and optimize the code, to make sure the system hardware and software systems achieve a fusion. The results show that, to ensure image quality and compression ratio of the premise, as far as possible to improve the encoding speed, the basic requirements of system have been meet.
Keywords/Search Tags:JPEG2000, static image compression, wavelet transform, DM642, arithmetic coding
PDF Full Text Request
Related items