The optimal color space enables advantageous smartphone-based colorimetric sensing

Smartphone-based colorimetric (bio)sensing is a promising alternative to conventional detection equipment for on-site testing, but it is often limited by sensitivity to lighting conditions. These issues are usually avoided using housings with fixed light sources, increasing the cost and complexity of the on-site test, where simplicity, portability, and affordability are a priority. In this study, we demonstrate that careful optimization of color space can significantly boost the performance of smartphone-based colorimetric sensing, enabling housing-free, illumination-invariant detection. We evaluated the quantification performance of smartphone-based colorimetry using monotonal shadings of colors with spectral compositions covering a wide range of visible spectra. The color coordinates were extracted from regions of interest (ROI) that were automatically selected using a homemade algorithm. Compared to absorbance-based models, smartphone-based colorimetry offered a broader measurement range with a comparable limit of detection. However, models based on RGB space proved highly sensitive to illumination changes, limiting their reliability. In contrast, the CIELAB color space—specifically the a∗ and b∗ chromatic coordinates—, exhibited inherent resistance to illumination changes. Our concept of equichromatic surfaces explains this inherent resilience to lightning variations, providing a theoretical basis for designing illumination-invariant optical (bio)sensors.