Design and Implementation of an Embedded NIOS II System for JPEG2000 Tier II Encoding

This paper presents a novel implementation of the JPEG2000 standard as a system on a chip (SoC). While most of the research in this field centers on acceleration of the EBCOT Tier I encoder, this work focuses on an embedded solution for EBCOT Tier II. Specifically, this paper proposes using an embedded softcore processor to perform Tier II processing as the back end of an encoding pipeline. The Altera NIOS II processor is chosen for the implementation and is coupled with existing embedded processing modules to realize a fully embedded JPEG2000 encoder. The design is synthesized on a Stratix IV FPGA and is shown to out perform other comparable SoC implementations by 39% in computation time. 1. Introduction One of the most recent image compression schemes, JPEG2000, offers a wide range of features and flexibility over the existing JPEG standard [1]. A block diagram of the JPEG2000 encoder is shown in Figure 1. The encoder consists of two main parts: the discrete wavelet transform (DWT) and the embedded block coding with optimal truncation (EBCOT) coder. The wavelet transform takes an image in the spatial domain and transforms it to the wavelet domain. The wavelet domain consists of a frequency representation with the addition of spatial information as well. Once the wavelet transform is completed, the coefficients are scalar quantized if lossy compression is chosen. The quantized wavelet coefficients are then entropy encoded using EBCOT, a two-tier coding algorithm which first divides each wavelet subband into code blocks (typically or ). EBCOT is composed of Tier I and Tier II encoders. Tier I produces independent embedded bitstreams for each code block using a context-based arithmetic encoder (MQ coder), the context for which is generated by the bit-plane coder. Tier II then reorders the individual compressed bitstreams and applies rate-distortion slope optimization to form the final JPEG2000 bitstream. Figure 1: Block diagram of JPEG2000 encoder. While JPEG2000 offers a number of improvements and additional features over JPEG and other image encoding standards, these benefits come with much greater computational cost. JPEG2000 is approximately 4 times more computationally expensive than the original JPEG [2]. Due to these high costs, it becomes impractical to utilize JPEG2000 in applications which require real-time processing of high-resolution images, such as wide area imagery or medical imagery. To solve this problem, developers continue to turn to hardware implementations to yield the throughput necessary to meet frame rates for high-resolution