We developed an automated method for removing binary background components from observed Raman spectra by tuning the scaling factors to seek the minimum lengths of the subtracted spectra. This method is effective, especially for large data including imaging data. For application, 400 Raman imaging spectra of a sliced cross section of a strand of gray human hair, fixed by glue on glass, were subjected to the proposed method by removing the glass and glue information. After the binary background removal, principal component analysis successfully detected small but important signals of tryptophan, which is peculiar to the hair cortex.

Spectral pre-treatments such as background removal from Raman big data are crucial to have a smooth link to advanced spectral analysis. Recently, we developed an automated background removal method, where we considered the shortest length of a spectrum by changing the scaling factor of the background spectrum. Here, we proposed a practical way to correct the systematic error caused by noise from measurement. This correction realized more effective and accurate automatic background removal.