以数字信号处理器 (DSP)为开发平台 ,实现了一种新的低存储量的图像编码算法。对如何在 DSP上实时实现运算量大、编码过程烦琐的小波零树编码算法进行了有益的尝试。对相关问题进行了深入的探讨 ,例如小波滤波器的选取 ,程序结构的并行优化 ,零树扫描过程的简化以及新算法“误差位数”的取值方案等等。大量实验证明 ,该文的实现结果满足一般视频编码的实时性要求 ,较传统的 JPEG图像编码标准有很大的进步 ,是可行的、高效的 DSP解决方案 With digital signal processor (DSP) as the development platform, a new low-memory image coding algorithm is implemented. This paper makes a useful attempt on how to realize wavelet zerotree coding algorithm with large computation amount and complicated coding process in real time on DSP. Such as the selection of wavelet filter, the parallel optimization of program structure, the simplification of zerotree scanning process and the scheme of “error bit number” of the new algorithm and so on. A large number of experiments show that the realization of the paper meets the real-time requirements of the general video coding, compared with the traditional JPEG image coding standard has made great progress, it is feasible and efficient DSP solution