As illustrated, fingerprint images 710, 720, 730, 740, 750, and 760 are combined to reduce, minimize, or cancel out the impact of the interference pattern caused by the non-uniform thickness of the contact layer. In one embodiment, fingerprint images 710, 720, 730, 740, 750, and 760 are combined by averaging fingerprint images 710, 720, 730, 740, 750, and 760 into a single output image. In the averaging, different weights may be applied to different images to account for differences in signal level and background level. In another embodiment, fingerprint images 710, 720, 730, 740, 750, and 760 are combined using convolution of fingerprint images 710, 720, 730, 740, 750, and 760 to generate a single output image. In some embodiments, only portions of images 710, 720, 730, 740, 750, and 760 exhibiting interference (e.g., above an interference threshold) are combined.

In one embodiment, a plurality of portions of an image is captured using the plurality of transmission frequencies, where some portions are captured at different transmission frequencies. The transmission frequencies of the portions are selected based on the thickness of contact layer 630 associated with the portions and the acoustic properties of the contact layer material, such as speed of sound. The plurality of portions captured is combined into a single output image, wherein the single output image reduces the impact of the interference pattern caused by the non-uniform thickness of the contact layer. It should be appreciated that each portion may have ToF and beamforming settings corresponding to the transmission frequency of the portion.