The most proper wavelet filters in low-complexity and an embedded hierarchical image compression structures for wireless sensor network implementation requirements

One major complication in implementing the discrete two-dimensional wavelet transform to a platform with limited resources is the need for huge memory. This paper addresses memory-efficient implementation of the wavelet-based image coding requirements. These requirements are usually distinct by resource-limited platforms such as tiny wireless sensors, which may build a wireless sensor network (WSN). Moreover, the bulky image data provided by the cameras combined with the network's resource constraints require discovering new means for data processing and communication. Image coding with scalar quantization on hierarchical structures of the transformed wavelet is considerably valuable and computationally simple. Typically, this is a case of set partitioning in hierarchical trees (SPIHT) a highly refined version of Embedded Zerotree Wavelet (EZW) structure that results from data similarity across different sub-bands. The paper deals with the effectiveness of an appropriate wavelet filter type that performs best results for SPIHT algorithm. The implementation of SPIHT structure based on the lifting scheme of wavelets is designed to compress several gray scale images with different information content in the MATLAB environment. Subjective and objective results are also evaluated and examined.