A Lossy Image Coding for Wireless Multimedia Sensor Networks

A simple, high-efficient and energy saving image coding scheme is proposed in this paper, which is suitable for wireless multimedia sensor networks. The scheme is based on lifting wavelet transform and SPIHT algorithm. To solve the very low executing efficiency problem of the combined algorithm, we make improvements in the two aspects. Firstly, by eliminating the computation and coding of the outmost high-pass coefficients, the energy consumption of the transform was reduced and the execution was speed up. Secondly, by using optimized quantization truncation pretreatment with high-pass subband coefficients, coding mass of small value coefficients of the high-pass subband was avoided, which further improves the executing efficiency. The simulation results show up to 90% reduction in the executing time was achieved by choosing the optimized truncation threshold, with the guarantee for the image quality of 34.5dB.