An Implementation Of A Real-Time Jpeg2000 Compressing System Based On DSP

JPEG2000, which is a new image compression standard, enables the achievement of higher image compression ratios than JPEG and also has superior features such as lower tile boundary noise and higher image quality. Further, it also features various powerful functions, such as highly hierarchical encoding, region-of-interest (ROI), lossless compression. Instead of using the discrete cosine transform (DCT) and Huffman coding used for JPEG, DWT and EBCOT are adopted. A huge amount of data processing must be carried out when performing DWT or EBCOT so that higher processing power is necessary than that required for JPEG. DWT also requires a huge memory capacity. Furthermore, costs of implementing the hardware capable of real-time JPEG2000 encode, specifically costs of circuitry and memory, are far greater than cost of JPEG hardware. Also regarding power consumption, a large circuit and memory as well as high operating frequency are obstacles to achieve a low power consumption JPEG2000 encoder. Therefore, it has been very difficult to develop a system using DSP of small size and low frequency JPEG2000-compliant hardware.This study introduces techniques and approaches capable of achieving a real-time JPEG2000 compressing system using DSP chips. Among several modules in JPEG2000 encoder, the lifting algorithm for discrete wavelet transform (DWT) and the embedded block coding with optimized truncation (EBCOT) comprise more than 85% of the encoding complexity. Thus, it is very important to design and optimize these two modules in order to increase the performance of the hardware implementation. We propose a five-DSP real-time parallel processing system using an high efficient memory management for discrete wavelet transform (DWT) and parallel-pass architecture for embedded block coding with optimized truncation (EBCOT). This system performs compression of 1392×1040 pixels monochrome images with the speed of 10 fps/camera of 4 digital still cameras and is proven to be a practical and efficient DSP solution. Key words: JPEG2000 DWT EBCOT DSP system Parallel processing...