Hello everybody, i'm working on a project for the develpment of a LABVIEW program for the discrimination of active principle fo pharmaceutical application. I'm using a NIR spectrometer for the acquisition of data. The data are stored in a 2D array. For every active principle i have a 2D array build as 50colum and 1024 row. every colum is one spectrum. what i would like to perform is a PCA analysis of such spectra in order to identify the different active principle. My idea is to acquire a reference spectra and then compare a test spectra. To do this i want to concatenate the 2D array of the two active principle and than perform PCA. I made it a small vi but i think is not correct. Someone has a suggestion or exists some example regarding such application?
I used the vi from the labview example "Principal component analysis" using my data instead of the example data. Anyway i think that such example is not adaptable to my data. attached to csv file from two different active principle. They are totally different active principle but from the scatter plot this different is not evident. The csv file is made of 50 row and 1024 colum. In this case every row is asingle spectrum.
I made 50 measurement for every actvive principle in the same point. From the plot the difference of the mean spectrum is evident. From the PCA such difference is not evident. I suppose i can't use the PCA vi from the Labview example. I hope now is more clear.
