Right now my colleague and I are working with a protocol for a set of photos that was taken using an already characterized camera (Sony A7). We want to be able to adapt the protocol to be used on a set of photos taken with a different, non-characterized camera. We know that the new version of the toolbox should allow us to proceed without knowing the spectral sensitivities of the camera. Can you point us in the right direction for the steps needed to use this non-characterized camera?
The user guide has all the info you need for this. You can either try to characterise the camera yourself (difficult – see the guide) or, if you’re using natural light only you can use an X-Rite colour checker to calibrate the camera. See here. You can’t use the colour-chart method with artificial light due to issues such as fluorescence.
Thanks so much for your reply. We were both able to create a cone mapping model using the color chart. However, I am now having issues with the process and am wondering if it is related to a MacOS update (my colleague still has 10.14.4, but I recently updated to 10.14.6). Following your tutorial, I am able to create an mspec of the color chart, measure the chart, and save the results as a csv file. But I then have issues when I go to generate the cone mapping model from the chart. Everything seems to work fine (I am able to select my results csv and choose my desired configuration), however it then gives me an error message “No window with the title “Results” found”. Do you have any idea what could be going on?
Thank you for your help!
This is a bug someone else has reported (possibly MacOS-related). As a temporary work-around, it should work if you measure the chart image and leave the window open (not ideal I know, but you can save the four measurement points as an ROI, and re-measure the image before making a new model).