Wiener filtering

[Artem_Y_Intel]

Hi, All.

I try to get IPP documentation example working.

Ipp32f pSrc[256*256];

Ipp32f pDst[256*256];

int srcStep = 256*sizeof(Ipp32f);

int dstStep = 256*sizeof(Ipp32f);

IppiSize roiSize = {256, 256};

IppiSize dstRoiSize = {256, 256};

IppiSize maskSize = {3, 3};

int pBufferSize;

int channels = 1;

unsigned int pSeed = 3;

IppiPoint anchor = {1, 1};

Ipp32f noise[1] = {0.5};

ippiImageJaehne_32f_C1R(pSrc, srcStep, roiSize); // source image

//source image + RandUniform noise

ippiAddRandUniform_Direct_32f_C1IR(pSrc, srcStep, roiSize, 0, 1, &pSeed);

ippiFilterWienerGetBufferSize( dstRoiSize, maskSize, channels, &pBufferSize);

Ipp8u* pBuffer = ippsMalloc_8u(pBufferSize);

ippiFilterWiener_32f_C1R( pSrc, srcStep, pDst, dstStep, dstRoiSize, maskSize, anchor, noise, pBuffer);

ippsFree(pBuffer);

After executing this code under Linux on Intel Xeon CPU L5408 @ 2.13GHz , IPP ver. 6.0.0.045 pDst buffer contains NAN values.

Is there anything I am doing wrong? Please, I need help.

Regards,

Artem