Characterization of a digital camera as an absolute tristimulus colorimeter

Abstract

An algorithm is proposed for the spectral and colorimetric characterization of digital still cameras (DSC) which allows them to be used as tele-colorimeters with CIE-XYZ color output, in cd/m 2 . The spectral characterization consists in the calculation of the color-matching functions from the previously measured spectral sensitivities. The colorimetric characterization consists in transforming the raw RGB digital data into absolute tristimulus values CIE-XYZ (in cd/m 2 ) under variable and unknown spectroradiometric conditions. Thus, in the first stage, a gray balance was applied over the raw RGB digital data to convert them into RGB relative colorimetric values. In the second stage, an algorithm of luminance adaptation versus lens aperture was inserted in the basic colorimetric profile. Capturing the ColorChecker chart under different light sources, and comparing the estimated XYZ data according to the developed color model in relation to the measured XYZ data (in cd/m 2 ) using a telespectroradiometer, we verified that the proposed characterization model may be broken down into two portions. Firstly, there is the basic colorimetric profile in combination with the new luminance adaptation algorithm. Secondly, there is the linear correction term due only to the mismatch of the color matching functions of the camera. Although the linear color correction term works relatively well, despite the imposed initial conditions (unknown spectral content of the scene), the separation of the proposed characterization model into two portions (raw and corrected performance) would allow the future comparison of various commercial cameras.

Software Development for Colourimetric and Spectral Data Processing: Pycolourimetry

“…The exposed methodology combines the direct method based on colourimetric measurements, and the indirect method using raw images with RGB information (Westland et al 2012;Martínez-Verdú et al 2003). The software developed in this study (pyColourimetry) covers each of the steps required for proper camera characterisation, and gives the user full control on the overall computation process.…”

“…Once the RGB to CIE XYZ transformation equations are obtained, a conventional digital camera could be used for rigorous colour determination as if it were a colourimeter (Martínez-Verdú et al 2003).…”

Concerning the calculation of the color gamut in a digital camera

“…To avoid these problems they instead chose to define first the gamut of the OCS in L*a*b* space. However, we have simulated the RGB responses for the optimal or MacAdam colors, which are on the boundary of the OCS, using our own characterization model 42 for digital cameras, mainly based on spectroradiometric measures. 43 They then obtain an approximation of the camera gamut by sampling this L*a*b* space and determining camera RGB values for each L*a*b* value sampled with a different camera model from that one of ours.…”

“…We consider that the illuminance of this hypothetic scene is 1000 ϫ l, obtained with a constant spectral radiance of 4.36 mW/sr m 2 , so the luminance of the adapting white is L E ϭ 1000/ cd/m 2 ϭ 318.31 cd/m 2 . With these lighting data, the camera exposure was set to N ϭ 4 and t ϭ 20 ms (offset value by manufacturer) and obtain the digital data RGB of each optimal color using our previously developed model 42,43 .…”

Conversion of a digital camera into a non-contact colorimeter for use in stone cultural heritage: The application case to Spanish granites

“…Several camera characterization techniques have been used 50 with the aim of developing a model (and estimating its parameters) for obtaining 51 L*a*b* color measurements from RGB digital values (e.g. [7][8][9][10][11][12][13][14][15]). In general, these 52 techniques can be divided into two categories: (1) spectral characterization, which 53 measures the three spectral-sensitivity functions for the red-green-blue (RGB) channels 54 and requires a monochromator and a radiance meter [16]; and (2) colorimetric 55 characterization, which involves mathematical transformations that yield the tristimulus 56 values from the digital values and which require use of a reference target that contains a 57 certain number of color samples.…”