Compression of ultraviolet–visible spectrum with recurrent neural network

Data compression method based on the recurrent neural network (RNN) of the dynamical system approach was proposed and applied to ultraviolet–visible (UV–VIS) spectra. RNN schemes with different network size were studied and their performance was evaluated by using both synthetic and experimental data. It was found that the storage space of the spectral information under study could be reduced significantly by using the proposed RNN method with quality spectra regenerated from the compressed data. Furthermore, the method was found to perform as good as the wavelet transform (WT) in data compression and in some cases, even better.