Quantitative colour evaluation of digital images based on illuminating spectral information

Abstract

We developed a quantitative colour appearance evaluation method of the images taken by a digital camera under an artificial lighting condition in the agricultural fields. By analysing the relationship between the colour appearance of the image of a standard colour chart with the plain red, green, and blue sections acquired by five kinds of digital cameras with the different specifications and the spectral feature information of the standard and artificial lighting sources, the photosensitive characteristics of the digital cameras were experimentally studied. As the results, the colour parameters of each colour section on the standard colour chart were expressed by a power function with two parameters for all tested digital cameras, and the colour parameters of each colour section on the virtual standard colour chart could be estimated. Additionally, the quantitative colour evaluation method under the natural lighting conditions developed in our previous study could be applied to the images taken under the illuminating conditions using an artificial lighting source 'three-wavelength high-frequency (Hf) fluorescent lamp'. Consequently, these results suggested that the developed method could be generally and widely adapted to the quantitative surface colour evaluation of agricultural products.