Reversible Data Hiding for Electrocardiogram Signal Based on Wavelet Transforms

A reversible watermarking algorithm with high data-hiding capacity has been developed for electrocardiogram (ECG) signal based on wavelet transforms. In electrocardiogram signal, the energy is concentrated in QRS complex waves. So the selection of wavelet coefficients for hiding should avoid making QRS complex waves distort obviously. The algorithm hides bits in the expansion of selected coefficients of high frequency sub-band of Harr wavelet transform based on lifting scheme. Due to a slightly modification on non QRS complex waves, the imperceptibility of the embedded watermark in the ECG signal is well guaranteed. The performance has been evaluated in terms of ECG signal distortion and embedding capacity. Experiment results demonstrate the efficiency of the watermarking scheme and show that the original ECG signal is reconstructed exactly after the retrieval of the watermark data.