STATISTICAL APPROACH TO COLOUR QUALITY

To compare the colour quality of different light sources is not an easy challenge. The International Commission on Illumination (Commission Internationale de l’Eclairge – CIE) proposed to use colour rendition index (CRI) in 1948. CRI is a quantitative measure of the ability of a light source to reproduce the colours of various objects faithfully in comparison with an ideal or natural light source. However this method is correct if relatively wide band light sources like incandesced and fluorescent lamps are considered. Several research studies have shown that lighting generated by LED based light sources with lower CRI index looks more attractive than light sources with higher CRI index. Today it is known that single index to describe light source colour quality is not enough, and CIE has came to a conclusion that “The CIE CRI is generally not applicable to predict the colour rendering rank order of a set of light sources when white LED light sources are involved in this set.” (CIE publ. 177:2007)

We suggest to use Statistical Colour Quality Metrics (SCQM), because this metric gives the most comprehensive information about light source ability to render colours. SCQM takes into account all colours listed in Munsell colour system (1269) and performs sorting of each colour sample in respect to the colour shift vector magnitude and direction in the CIELAB colour space. If the vector length does not exceed the dimension of interpolated MacAdam ellipse multiplied by factor three, then Colour Fidelity Index (CFI) is scored meaning that the colour sample is rendered with high fidelity and a human with statistical properties of vision will not notice the colour shift. If the colour shift vector exceeds the size of thee MacAdam ellipses and is directed to the increased or decreased chroma and/or to the hue change, then Colour Saturation Index (CSI) or Colour Dulling Index (CDI) and/or Hue Distortion Index (HDI) are scored respectively. In addition, if the lightness of the sample is changed more that by 2%, then Lightness Distortion index is scored. After the sorting of all Munsell samples (1269), the colour rendition indexes are expressed in easy understandable percentage value, meaning that under tested illumination CFI[%] of colours are rendered with high fidelity, CSI[%] of colours are saturated, CDI[%] of colours are dulled, and HDI[%] and LDI[%] of colours have distorted hue and lightness respectively.

The most commonly used white light sources can be grouped into three groups namely High fidelity, Colour Saturating and Colour dulling. Please take attention that LED clusters of four initial colours (Red, Amber, Green and Blue – RAGB) can be found in all three groups depending on the partial intensities of each channel and resulting Spectral Power Distribution.

For more detailed description of the SCQM metrics please refer to the following scientific papers:

• A. Žukauskas, R. Vaicekauskas, F. Ivanauskas, H. Vaitkevičius, P. Vitta, and M. S. Shur “Statistical Approach to Color Quality of Solid-State Lamps,” IEEE Sel. J. Sel Top. Quantum Electron. 15 (4), pp 1189-1198, (2009). DOI: 10.1109/JSTQE.2009.2034587
• A. Žukauskas, R. Vaicekauskas, and M. Shur “Solid-state lamps with optimized color saturation ability,” Opt. Exp. 18 (3), pp. 2287-2295 (2010). (Free access)
• A. Žukauskas, R. Vaicekauskas, and M. Shur “Colour-rendition properties of solid-state lamps,” J. Phys. D. Appl. Phys. 43 (35), 354006 (2010). DOI: 10.1088/0022-3727/43/35/354006
• A. Žukauskas, R. Vaicekauskas, and M. Shur “Color-dulling solid-state sources of light,” Opt. Exp. 20 (9), pp. 9755-9762 (2012). (Free access)

COLOUR RENDITION ENGINE

(PCT patent publication WO/2013/009157)

Ability to saturate or dull colours

Tetrachromatic solid state light source have an ability to saturate or dull colours depending on individual needs. It is extremely important, when representing faded objects. Furthermore objects with saturated colours attract more attention which is very important in retail business. Some objects require dulled illumination for example ecclesiastical and spelaean art. Commercial light sources allow changing the colour saturation-dulling ratio with wireless software control – it is enough to take smart mobile device (smartphone or tab) or PC and setup lighting parameters by pushing a few buttons. Example bellow illustrates how the appearance of the fruit scene can be changed (be aware that the camera and computer monitor is not capable to reproduce all colours in a right way).

Lighting with preferred parameters

Light sources with high colour rendition index R_a or with high colour fidelity index CFI are not always the most desired by the humans. Research shows, that people prefer a little bit saturated lighting in compare with natural illumination.

More than 250 subjects were asked to adjust saturation-duling factor of a tetrachromatic LED cluster (colour rendition engine) in order to perform a trade-off between colour saturation and colour dulling effects. As an object was presented several paintings, one of them is shoved above – colourful painting with dominant fire theme. The results showed that the average preference was given to somewhat saturating lighting rather than to high-fidelity colour rendition. Similar settings of proffered illumination were observed for other scenes (fruits, paintings) also.

Ability to shift colour gamut

Objects, especially artworks, under inappropriate illumination change colour gamut due to the photochemical damage. Until now the restoration of the paintings and other aged artworks was the only way to restore colours and entire colour gamut. Our tuneable illumination technology allows for precise and mathematically proven shift of colour gamut in order to obtain the desired result. For example the paintings with aged and yellowed varnish can be restored in real time by shifting the gamut towards the bluish region of the colour space. Below it is shown the colour gamut shift for the brownish picture in the CIELAB colour space. Each point represent the pixel of the painting recorded by the imaging colour photometer.

More detailed description of the Colour Rendintion Engine can be found in this publication

• Žukauskas, R Vaicekauskas, P. Vitta, A. Tuzikas, A. Petrulis, and M. Shur ” Color rendition engine,” Opt. Exp. 20 (5), pp. 5356-5367 (2012). (Free access)