OCR
a CHARISM A I 7 Mes T D65 T RGB XYZ XYZ This procedure has been incorporated into a post-processing workspace for UV-induced luminescence images using nip2. It applies the colour calibration matrix M (determined for the calibration of VIS images, captured with an illuminant of colour temperature T), to the UVL image acquired with a white balance set to D65 and imaged with the camera in RGB colour space. This creates an image in ‘D65’, which requires a colour temperature transform, MT, to compensate for the difference in colour temperature between D65 and T, and move the result back to the equivalent CIE XYZ colour space under a D65 illuminant. In the post-processing workspace this approach to the white-point adjustment has been developed into a slider, so that the user can select the calibration illuminant colour temperature T (used to acquire the VIS image and to determine M), from a source with known colour temperature e.g. a D50 flash, or as measured with a colour temperature meter (see Chapter 3). However, the approximation assumes that the changes in colour temperature between T and D65 will not be very large. This is likely to be the case, since the overwhelming choice of illuminant used in the acquisition of VIS images are flash lights which are approximately D50, and hence close to D65. Trials have shown that for other illuminants such as tungsten lamps (D32), the change in colour temperature is too great and the approximation breaks down.*? Currently the above method has only been applied to UV-induced luminescence images but not to visible-induced infrared luminescence images as these are monochrome. The method could possibly be applied to visible-induced visible images but the post-processing of this image type is outside the scope of this work. The approach will be integrated into a workflow for the development of the post-processing software addressing the correction of luminescence images (see Chapter 3). The optimisation of experimental procedures, as well as the data acquisition requirements for post-processing, are discussed in Chapter 2. Version No. 1.0 30 Date : 14/10/2013