This forum is disabled, please visit https://forum.opencv.org
 | 1 | initial version |
This is what I ended up using:
double Spectrum[N_SPECT][PIXEL_COUNT]; //define contiguous memory block
[...]
DLL_Get_Spectra(Spectrum[SpectrumCount], &CaseTemp);// Grab data into Spectrum[][] one row at a time
[...]
cv::Mat_<double> Image(SpectrumCount, PIXEL_COUNT, &Spectrum[0][0]);//build Mat around Spectrum[][]
It works quite well. I decided to use Mat_<double> instead of Mat(,,CV_64F) to be absolutely sure that I get the pixel type right - Microsoft is very fuzzy on the exact size of a double. Why the spectrometer is returning 16 bit integers as double in the first place is beyond me.
guy
| | 2 | No.2 Revision |
This is what I ended up using:
double Spectrum[N_SPECT][PIXEL_COUNT]; //define contiguous memory block
[...]
DLL_Get_Spectra(Spectrum[SpectrumCount], &CaseTemp);// Grab data into Spectrum[][] one row at a time
[...]
cv::Mat_<double> Image(SpectrumCount, PIXEL_COUNT, &Spectrum[0][0]);//build Mat around Spectrum[][]
It works quite well. I decided to use Mat_<double> instead of Mat(,,CV_64F) to be absolutely sure that I get the pixel type right - Microsoft is very I was somewhat fuzzy on the exact size of a double. Why the spectrometer is returning 16 bit integers as double in the first place is beyond me.
guy