Color calibration and fusion of lens-free and mobile-phone microscopy images for high-resolution and accurate color reproduction

• Published in: Scientific reports Vol. 6; no. 1; p. 27811
• Main Authors: Zhang, Yibo, Wu, Yichen, Zhang, Yun, Ozcan, Aydogan
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 10.06.2016; Nature Publishing Group
• Subjects: 639/624/1075; 639/624/1107/510; Calibration; Colorization; Humanities and Social Sciences; Microscopy; multidisciplinary; Science; Semiconductors; Sensors; Wavelengths
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/srep27811

Lens-free holographic microscopy can achieve wide-field imaging in a cost-effective and field-portable setup, making it a promising technique for point-of-care and telepathology applications. However, due to relatively narrow-band sources used in holographic microscopy, conventional colorization methods that use images reconstructed at discrete wavelengths, corresponding to e.g., red (R), green (G) and blue (B) channels, are subject to color artifacts. Furthermore, these existing RGB colorization methods do not match the chromatic perception of human vision. Here we present a high-color-fidelity and high-resolution imaging method, termed “digital color fusion microscopy” (DCFM), which fuses a holographic image acquired at a single wavelength with a color-calibrated image taken by a low-magnification lens-based microscope using a wavelet transform-based colorization method. We demonstrate accurate color reproduction of DCFM by imaging stained tissue sections. In particular we show that a lens-free holographic microscope in combination with a cost-effective mobile-phone-based microscope can generate color images of specimens, performing very close to a high numerical-aperture (NA) benchtop microscope that is corrected for color distortions and chromatic aberrations, also matching the chromatic response of human vision. This method can be useful for wide-field imaging needs in telepathology applications and in resource-limited settings, where whole-slide scanning microscopy systems are not available.