Wavelet-based morphological correlation

A wavelet-based morphological correlation (WBMC) is proposed as a new architecture to improve the properties of the classical morphological correlation (MC). For the WBMC, a dilated wavelet intensity function is introduced to filter the joint power spectrum (JPS) of the MC before final inverse Fourier transform. Computer simulation results show that, as compared with the linear correlation (LC), the conventional MC and the joint wavelet transform correlation (JWTC), the WBMC provides better discrimination capability with sharp and unmistakable correlation signal and its performance metrics are more stable under input outlier noise (salt-and-pepper noise). Although the WBMC loses illumination-invariance when input illumination factor is larger than unity, considerable discrimination capability is still maintained.