Convert Bayer RGGB to CV iplimage (RGB)

Question

currently I'm trying to convert the Bayer RGGB data to an iplimage. I thought cvtColor might work but it requires "mat" instead of iplimage.

cvtColor(img->imageData, tmpimageData, CV_BayerBG2BGR, 0);

Is there a workaround, maybe you can convert Bayer RGGB to mat and then convert mat to iplimage? I'm really stuck with this problem, any help is greatly appreciated!

score 1 · Answer 1 · answered Jan 07 '14 at 18:19

1

OpenCV can easily convert between Mat and Iplimage. E.g., if img is an IplImage*:

Mat m(img);
Mat result;
cvtColor(m, result, CV_BayerBG2BGR, 0);
IplImage ipl_result  = IplImage(result);

See Mat::Mat(const IplImage* img, bool copyData=false) and Mat::operator IplImage()

answered Jan 07 '14 at 18:19

Bull

11,771
9
42
53

Well that's quite usefull ^^ I guess i can use img again, and don't need to create the new IplImage "ipl_result"! – Guntram Jan 13 '14 at 11:30

score 0 · Answer 2 · answered Feb 18 '14 at 18:17

Unfortunately I didn't figure out why cvtColor didn't work, but nonetheless I found a why to convert BayerRGGB to RGB with my SDK.

It turned out that the matrix-vision sdk I was working with provided a built in solution. The first thing I needed to do was to change PixelFormat to BayerRG8 in order to get 8 bit Image resolution. After that, I managed to decode Bayer RGGB to RGB by writing idpfRGB888Packed into ImageDestination.pixelFormat.

Also, I was able to do my conversion in a very primitive way, which also got the job done but required way too much cpu time. I looped my imageData array and took out every fourth pixel.

for (int count_small = 0, count_large = 0; count_large < 1000; count_small += 3, count_large +=4)
{
  dest[count_small] = source[count_large];
  dest[count_small+1] = source[count_large+1];
  dest[count_small+2] = source[count_large+2];
}

Not clean but it got the job done.

Convert Bayer RGGB to CV iplimage (RGB)

2 Answers2