Image Compression Techniques in Wireless Sensor Networks: A Survey and Comparison

There is continuous intensive research on image compression techniques in wireless sensor networks (WSNs) the literature. Some of WSNs that exist literature include discrete cosine transform (DCT), waveform transforms (DWT), set partitioning a hierarchical tree (SPIHT), and embedded zero wavelet (EZW) coding. Research necessitated by need to improve energy efficiency nodes WSNs’ lifetimes without compromising quality reconstructed data. Several approaches have been developed centered around other factors trying limit consumption nodes. Most these do not provide error-bound mechanism balances rate distortion image. Therefore, this paper, review analysis are conducted. Available were then classified according technique adopted, their strengths weaknesses highlighted. In addition, rate-distortion balanced data algorithm with error bound based artificial neural (ANN) form an autoencoder (AE) was coded simulated MATLAB before being evaluated compared conventional approaches. The experimental results show has less root mean square (RMSE) higher coefficient determination (R2) values variable ratios as Principal Component Analysis (PCA), Discrete Cosine Transform, Fast Fourier Transform (FFT) when using Grand-St-Bernard metrological dataset. Furthermore, it presented RMSE, ratio Lightweight Temporal Compression (LTC) bounds LUCE found presents better fidelity mechanism. Moreover, analyzed significant approach balancing through its