Title :
Fast JPEG Huffman Table Restoring and Decoding for Embedded DSP Implementations
Author :
Yang, Pei-Man ; Yang, Jen-Kuei ; Wang, Wen-Shan ; Tseng, Shau-Yin
Author_Institution :
SoC Technol. Center, Ind. Technol. Res. Inst., Hsinchu, Taiwan
Abstract :
The contents of the Huffman tables for JPEG coding might varied from one image to another. To ensure that the JPEG decoder uses the same Huffman tables as the encoder did, the JPEG standard suggests a method for encoder to downsize the Huffman tables and save them in file header so that the decoder can restore them from the header. Table restoration is time consumed because of the huge AC Huffman table size. It is a challenge for the embedded devices to make this real-time task due to lack of hardware support and limited memory space. This paper introduces a fast approach to restore the Huffman tables and reduce the table size. According to the experiment results run in ARM7TDMI, the proposed approach can present, compared with the leading-bit search, about 80% size down for table buffers and about 3 times of speedup for the table building.
Keywords :
Huffman codes; decoding; digital signal processing chips; image coding; image restoration; JPEG coding; JPEG decoder; digital signal processing; embedded DSP; fast JPEG Huffman table restoration; file header; image coding; Arithmetic; Code standards; Decoding; Digital signal processing; Entropy; Hardware; Image coding; Image restoration; Transform coding; Video coding;
Conference_Titel :
Multimedia, 2009. ISM '09. 11th IEEE International Symposium on
Conference_Location :
San Diego, CA
Print_ISBN :
978-1-4244-5231-6
Electronic_ISBN :
978-0-7695-3890-7
DOI :
10.1109/ISM.2009.18