Recent studies have reported wing interference patterns (WIPs), which reflect the microstructure of the wing, for small insects belonging to the Diptera and Hymenoptera orders. WIPs have been evaluated using RGB or multispectral images, but in contrast to these approaches, hyperspectral images allow a more detailed analysis of spectral variation, which may not be captured by RGB or multispectral images. Here, I investigated the WIPs of 12 Drosophila species using hyperspectral images. The average spectrum was calculated for each of the six compartments of the wing region and for the entire wing, including all six compartments. This information was used to evaluate sexual and interspecific differences in the WIPs of 12 Drosophila species. In addition, the possibility of species discrimination based on WIPs was explored using the random forest machine learning algorithm. The present study demonstrates significant sex and interspecific differences in WIPs for each of the six compartments of the wing regions as well as for the entire wing region. The results of the random forest machine learning algorithm suggested the possibility of species identification based on WIPs. Hyperspectral images, which store data fromlight that is spectralized into dozens or more bands for each wavelength range, allow a more detailed analysis of spectral variation. Here, wing interference patterns (WIPs) of 12 Drosophila species were investigated using hyperspectral images. The present study demonstrates significant sex and interspecific differences in WIPs, and suggests the possibility of species identification based on WIPs.