寻找与英特尔IPP进行FFT卷积的工作示例

Question

现在正试图绕过IPP。有没有人有任何工作的C / C ++代码用于对图像进行FFT卷积？

Answer 1

Example from here：

/* Sample C-code snipet for Example 1 using Intel® Integrated Performance Primitives (Intel® IPP): */

/* allocate and initialize specification structures */
ippsFFTInitAlloc_C_32fc(&FFTspec1_p, order, IPP_FFT_DIV_FWD_BY_N, ippAlgHintFast);
ippsFFTGetBufSize_C_32fc(FFTspec1, &BufSize);
Buf1_p = (Ipp8u *) ippsMalloc_32sc(BufSize*sizeof(Ipp8u));

// ...

/* compute in-place FFTs of input sequences*/
ippsFFTFwd_CToC_32fc_I(x_p, FFTspec1_p, Buf1_p);
ippsFFTFwd_CToC_32fc_I(y_p, FFTspec1_p, Buf1_p);

/* perform complex multiplication and inverse FFT*/
ippsMul_32fc( x_p, y_p, o_p, veclength);
ippsFFTInv_CToC_32fc_I(o_p,FFTspec1_p, Buf1_p);

// ...

/* free specification structures */
ippsFFTFree_C_32fc( FFTSpec1_p);
ippsFree(Buf1_p);

Answer 2

我也在研究图像的FFT卷积。我的询问表明以“ipps”开头的ipp函数是针对信号处理的。以“ippi”开头的Ipp函数针对图像处理。有关工作示例，请参阅

的第597页

http://download-software.intel.com/sites/products/documentation/doclib/ipp_sa/71/ipp_manual/ippi.pdf