From the plots above, it is clearly observable that the green channel is somewhat linearly correlated with the dilution factor. Both Fig. \ref{365956} and Fig. \ref{240772}  agree with that. However, the plots of the blue and red channel are completely different in Fig. \ref{365956} and Fig. \ref{240772}.
Therefore, I worked on extracting channel values from the image itself (rather than using the provided excel sheet). For this, I wrote a script in python that automatically extracts the pixel values from reference and sample zone, and stores them in an array. To test different algorithms, we can simply re-use this array without having to re-extract the values.