R-D Optimization Of Tier₂ Encoder In JPEG2000

Image compression is the key for the storage and transmission of the huge data in the modern image processing. The traditional still image compression techniques such as the JPEG standard could not meet the rapidly increased demand for the high quality low bit-rate images in various emerging applications today. The latest still image compression standard JPEG2000 adopts Discrete Wavelet Transform (DWT) and Embedded Block Coding with Optimized Truncation (EBCOT), which achieves much better coding efficiency and image quality than the traditional methods. With its excellent performance and prominent features, JPEG2000 will become the main-stream in the image compression field in future;Under the given distortion conditions, the still image coding Rate-Distortion theories discuss the max coding bit-rate. Unde the given coding bit-rate, the max coding distortion decrease can be calculated mith Rate-Distortion algorithms.The first coding bits have more information and reflecet the image outlines, the last coding bits reflect the image details. So the Rate-Distortion theory is very important for the embedded still image coding.This article discussed the basic principle of the still picture rate distortion theory in detail with the base of the still image compression standard JPEG2000. Based on the analysis of the main obstacle of embedded still image coding this paper focus on the coding order with Rate_Distortion Optimized, subband bit allocations and quantization step sizes. Proposed two R_D optimization algorithms: Based on the steepest R_D slope coding order and based on Lagrange Multiplier, and both of the two algorithms are success applied in the image coding hardware system. The simulation result indicated that the Codingspeed of these algorithms code speed is very quick, the realization is simple and the code effect is better.