Validade e repetibilidade da avaliação de cor com fotografias digitais

Abstract

Objective: This study aimed to verify the validity of color measurement with a digital camera with different white balance (WB) setups, as well as to determine its repeatability. Methods: Twelve shade tabs (B1-B4, A1-A4, C1-C3) of a „Vitapan Classical‟ shade guide had their colors measured with clinical spectrophotometer, and with digital images to obtain the CIE L*a*b* color parameters. For each shade tab, ten (10) images were taken with a camera with customized white balance (CWB), and other 10 with automatic white balance (AWB) setups. To determine validity, the Kappa coefficient was calculated considering the twelve shade tabs both individually and divided in four groups according chroma. Sensitivity and specificity tests were also carried out. The Friedman test was performed to verify the repeatability of the images. Results: „Fair agreement‟ for images with AWB and „substantial agreement‟ for images with CWB were found. Considering the four color categories, images with AWB showed a „moderate agreement‟, while those with CWB still reported a „substantial agreement‟ but with a higher kappa value. The CWB setup had also the best results of sensitivity (from 63.3% for darker-intermediate colors to 100% for lighter colors), and specificity (from 88.8% for darker colors to 98.8% for lighter-intermediate colors). The images also had a great repeatability for both CWB and AWB setups (p>0.05). Conclusion: Digital images proved to be a valid instrument for color measurement, with an excellent repeatability, as long as the images are taken by a camera with customized white balance setup. Thus, that can be a reliable alternative to the use of spectrophotometers.