In seismic mapping of reservoirs, it is usually difficult to identify a single layer in a thin interbed formation. The main reason for this is interference of adjacent layers in the thin beds. We have developed a slice method basis spectral decomposition workflow to decrease this interference rely on the good vertical resolution of the thin layers, which is not always achievable. To overcome this, we have developed two frequencies at which the thin-bed response is weakest at a fixed time. These are the low minimum interference (LMI) and high minimum interference frequencies. The LMI frequency is useful for thin-bed detection when the thin bed cannot be resolved vertically in the seismic profile. The LMI frequency is usually less than the dominant frequency. However, the thin bed can still be identified on the seismic slice at this frequency. In practice, the LMI frequency can be estimated from the time-frequency spectrum and used in spectral decomposition using wavelet transformation to generate the amplitude slices for various frequencies. The amplitude slicing method was verified by case studies using synthetic and real data from a seismic exploration site in the Junggar Basin, China. The results indicate that this methodology is feasible and effective in identifying oil reservoirs in the thin interbeds strata.