Hello Jolyon, and thank you for developing this amazing piece of software!
I consistently get the same error when trying to generate a cone mapping model from the measurements of a x-rite mini colour checker. I tried to use different cameras (each one with its specific custom linearisation model which seems to work properly) and different pictures of the checker, but the same error always shows up.
The following is part of the error message I’m getting:
ImageJ 1.52k; Java 1.8.0_172 [64-bit]; Mac OS X 10.14.5; 1143MB of 3000MB (38%)
warning: [options] bootstrap class path not set in conjunction with -source 1.6
/Applications/ImageJ.app/plugins/Cone Models/NikonCoolpixAW130_CIE_XYZ_1931_D65_.java:26: error: not a statement
/Applications/ImageJ.app/plugins/Cone Models/NikonCoolpixAW130_CIE_XYZ_1931_D65_.java:38: error: illegal start of expression
= (float) stack.getPixels(1);
/Applications/ImageJ.app/plugins/Cone Models/NikonCoolpixAW130_CIE_XYZ_1931_D65_.java:44: error: ‘)’ expected
X[i] = (float) (NaN [i]+( [i]*NaN)+(visibleR[i]*NaN)+(visibleG[i]*NaN)+(visibleB[i]*NaN)+( [i]*visibleR[i]*NaN)+( [i]*visibleG[i]*NaN)+( [i]*visibleB[i]*NaN)+(visibleR[i]*visibleG[i]*NaN)+(visibleR[i]*visibleB[i]*NaN)+(visibleG[i]*visibleB[i]*NaN));
Am I missing something?
Thanks for the help!
The code is showing you have a load of NaNs where there should be numbers. So evidently something has gone wrong with either the measurement of chart values, or the loading of chart values from the file, or the labels of the camera channels.
When you measure your photograph of the chart do you get “Red” “Green” and “Blue” listed as the channels?
I’m having the same problem currently. Have either of you figured out the source of this error? Thanks!
Yes I did. The error is caused by the Result file being saved with row numbers included. I did not figure out a way to save the result file without them yet.
However, a way around the problem is to measure the colour chart and then directly generate a cone mapping model without closing the result file. It is tedious because you need to remeasure the chart every time you want to model a different visual system, but it is the only solution I found for MacOS right now.
Hope it helps!
Thanks for the update Matteo. I think I might have to bypass ImageJ’s reading of results files and write some code to parse .csv files manually.
I’m very excited to have started using your software.
I’m also having the same issues as the others on this post. I followed Matteo’s suggestion and kept my results.csv file open while I try to genereate a cone mapping model but I still have the same error reading. Have you or anyone else managed to come up with a different solution? I’m not sure if I’m doing something wrong. Thanks!
I’ve attempted to fix this bug with the latest version. Try downloading the latest version and see whether that works!
Thanks for the reply! I did realise about the NaNs but could not figure out why they appear.
When I measure the chart I do get the three channels listed (“visible:R:Normalised”, “visible:G:Normalised”, and “visible:B:Normalised”). Also the loading of the chart values from the “Results” csv file seems to work fine, and the file appears identical to the one you show in the tutorial video..