An Efficient Architecture for 3-D Discrete Wavelet Transform

This paper presents an architecture of the lifting-based running 3-D discrete wavelet transform (DWT), which is a powerful image and video compression algorithm. The proposed design is one of the first lifting based complete 3-D-DWT architectures without group of pictures restriction. The new computing technique based on analysis of lifting signal flow graph minimizes the storage requirement. This architecture enjoys reduced memory referencing and related low power consumption, low latency, and high throughput compared to those of earlier reported works. The proposed architecture has been successfully implemented on Xilinx Virtex-IV series field-programmable gate array, offering a speed of 321 MHz, making it suitable for real-time compression even with large frame dimensions. Moreover, the architecture is fully scalable beyond the present coherent Daubechies filterbank (9, 7).