<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic even mask ippiFilterColumn_32f_C1R in Intel® Integrated Performance Primitives</title>
    <link>https://community.intel.com/t5/Intel-Integrated-Performance/even-mask-ippiFilterColumn-32f-C1R/m-p/977286#M21092</link>
    <description>1. I want to calculate the average of y.&lt;BR /&gt;2. When the size of the mask is an even number, it is incorrect.&lt;BR /&gt;   (Pen3 i820 WinXpSp1 IPP4.1.22)&lt;BR /&gt;3. Is there something alternative proposal?&lt;BR /&gt;&lt;BR /&gt;// sample code&lt;BR /&gt;void main(void)&lt;BR /&gt;{&lt;BR /&gt;    Ipp32f *pSrc; int SrcW, SrcH, SrcS;&lt;BR /&gt;    Ipp32f *pDst; int DstW, DstH, DstS;&lt;BR /&gt;    Ipp32f *pMsk; int MskSize;&lt;BR /&gt;    Ipp32f *pTmp; int yAnchor, x;&lt;BR /&gt;&lt;BR /&gt;    IppStatus Status;&lt;BR /&gt;    IppiSize SrcSize, DstSize;&lt;BR /&gt;&lt;BR /&gt;    SrcW = 5; SrcH = 4;&lt;BR /&gt;    pSrc = ippiMalloc_32f_C1(SrcW, SrcH, &amp;amp;SrcS);&lt;BR /&gt;    SrcSize.width = SrcW; SrcSize.height = SrcH;&lt;BR /&gt;    Status = ippiSet_32f_C1R(1.0f, pSrc, SrcS, SrcSize);&lt;BR /&gt;&lt;BR /&gt;    MskSize = SrcH;&lt;BR /&gt;    pMsk = ippsMalloc_32f(MskSize);&lt;BR /&gt;    Status = ippsSet_32f(1.0f / MskSize, pMsk, MskSize);&lt;BR /&gt;&lt;BR /&gt;    DstW = SrcW; DstH = 1;&lt;BR /&gt;    pDst = ippiMalloc_32f_C1(DstW, DstH, &amp;amp;DstS);&lt;BR /&gt;&lt;BR /&gt;    yAnchor = (MskSize - 1) / 2;&lt;BR /&gt;    pTmp = pSrc + (SrcS / sizeof(Ipp32f)) * yAnchor;&lt;BR /&gt;    DstSize.width = DstW; DstSize.height = DstH;&lt;BR /&gt;    Status = ippiFilterColumn_32f_C1R(&lt;BR /&gt;        pTmp, SrcS, pDst, DstS, DstSize, &lt;BR /&gt;        pMsk, MskSize, yAnchor);&lt;BR /&gt;    printf("Status = %d
", Status);&lt;BR /&gt;    pTmp = pDst;&lt;BR /&gt;    for ( x = 0; x &amp;lt; DstW; x++ )&lt;BR /&gt;    {&lt;BR /&gt;      printf("%f
", *pTmp); pTmp++;  &lt;BR /&gt;    }&lt;BR /&gt;&lt;BR /&gt;    ippiFree(pDst);&lt;BR /&gt;    ippsFree(pMsk);&lt;BR /&gt;    ippiFree(pSrc);&lt;BR /&gt;&lt;BR /&gt;    printf("push any key
"); getch();&lt;BR /&gt;}</description>
    <pubDate>Mon, 05 Sep 2005 15:32:30 GMT</pubDate>
    <dc:creator>seiji-torigoe</dc:creator>
    <dc:date>2005-09-05T15:32:30Z</dc:date>
    <item>
      <title>even mask ippiFilterColumn_32f_C1R</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/even-mask-ippiFilterColumn-32f-C1R/m-p/977286#M21092</link>
      <description>1. I want to calculate the average of y.&lt;BR /&gt;2. When the size of the mask is an even number, it is incorrect.&lt;BR /&gt;   (Pen3 i820 WinXpSp1 IPP4.1.22)&lt;BR /&gt;3. Is there something alternative proposal?&lt;BR /&gt;&lt;BR /&gt;// sample code&lt;BR /&gt;void main(void)&lt;BR /&gt;{&lt;BR /&gt;    Ipp32f *pSrc; int SrcW, SrcH, SrcS;&lt;BR /&gt;    Ipp32f *pDst; int DstW, DstH, DstS;&lt;BR /&gt;    Ipp32f *pMsk; int MskSize;&lt;BR /&gt;    Ipp32f *pTmp; int yAnchor, x;&lt;BR /&gt;&lt;BR /&gt;    IppStatus Status;&lt;BR /&gt;    IppiSize SrcSize, DstSize;&lt;BR /&gt;&lt;BR /&gt;    SrcW = 5; SrcH = 4;&lt;BR /&gt;    pSrc = ippiMalloc_32f_C1(SrcW, SrcH, &amp;amp;SrcS);&lt;BR /&gt;    SrcSize.width = SrcW; SrcSize.height = SrcH;&lt;BR /&gt;    Status = ippiSet_32f_C1R(1.0f, pSrc, SrcS, SrcSize);&lt;BR /&gt;&lt;BR /&gt;    MskSize = SrcH;&lt;BR /&gt;    pMsk = ippsMalloc_32f(MskSize);&lt;BR /&gt;    Status = ippsSet_32f(1.0f / MskSize, pMsk, MskSize);&lt;BR /&gt;&lt;BR /&gt;    DstW = SrcW; DstH = 1;&lt;BR /&gt;    pDst = ippiMalloc_32f_C1(DstW, DstH, &amp;amp;DstS);&lt;BR /&gt;&lt;BR /&gt;    yAnchor = (MskSize - 1) / 2;&lt;BR /&gt;    pTmp = pSrc + (SrcS / sizeof(Ipp32f)) * yAnchor;&lt;BR /&gt;    DstSize.width = DstW; DstSize.height = DstH;&lt;BR /&gt;    Status = ippiFilterColumn_32f_C1R(&lt;BR /&gt;        pTmp, SrcS, pDst, DstS, DstSize, &lt;BR /&gt;        pMsk, MskSize, yAnchor);&lt;BR /&gt;    printf("Status = %d
", Status);&lt;BR /&gt;    pTmp = pDst;&lt;BR /&gt;    for ( x = 0; x &amp;lt; DstW; x++ )&lt;BR /&gt;    {&lt;BR /&gt;      printf("%f
", *pTmp); pTmp++;  &lt;BR /&gt;    }&lt;BR /&gt;&lt;BR /&gt;    ippiFree(pDst);&lt;BR /&gt;    ippsFree(pMsk);&lt;BR /&gt;    ippiFree(pSrc);&lt;BR /&gt;&lt;BR /&gt;    printf("push any key
"); getch();&lt;BR /&gt;}</description>
      <pubDate>Mon, 05 Sep 2005 15:32:30 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/even-mask-ippiFilterColumn-32f-C1R/m-p/977286#M21092</guid>
      <dc:creator>seiji-torigoe</dc:creator>
      <dc:date>2005-09-05T15:32:30Z</dc:date>
    </item>
    <item>
      <title>Re: even mask ippiFilterColumn_32f_C1R</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/even-mask-ippiFilterColumn-32f-C1R/m-p/977287#M21093</link>
      <description>&lt;DIV&gt;Hi,&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;there is answer from our expert:&lt;/DIV&gt;
&lt;DIV&gt;&lt;FONT color="#000080" size="2"&gt;
&lt;P&gt;seemsyou forget (or misread) that arbitrary kernels for filtering operations are used in inverse order (the same as for FIR), therefore anchor should be treated in inverse order too: &lt;/P&gt;&lt;/FONT&gt;&lt;FONT face="Courier New CYR" size="2"&gt;
&lt;P&gt;pTmp = pSrc + (SrcS / &lt;/P&gt;&lt;/FONT&gt;&lt;FONT face="Courier New CYR" color="#0000ff" size="2"&gt;sizeof&lt;/FONT&gt;&lt;FONT face="Courier New CYR" size="2"&gt;(Ipp32f)) * ( MskSize - yAnchor - 1 );&lt;P&gt;&lt;/P&gt;&lt;/FONT&gt;&lt;FONT color="#000080" size="2"&gt;
&lt;P&gt;The best way for calculating average is FilterBox or SumColumn function  they are faster because dont use muls.&lt;/P&gt;&lt;/FONT&gt;&lt;/DIV&gt;
&lt;DIV&gt;Regards,&lt;/DIV&gt;
&lt;DIV&gt; Vladimir&lt;/DIV&gt;</description>
      <pubDate>Thu, 08 Sep 2005 23:43:59 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/even-mask-ippiFilterColumn-32f-C1R/m-p/977287#M21093</guid>
      <dc:creator>Vladimir_Dudnik</dc:creator>
      <dc:date>2005-09-08T23:43:59Z</dc:date>
    </item>
    <item>
      <title>Re: even mask ippiFilterColumn_32f_C1R</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/even-mask-ippiFilterColumn-32f-C1R/m-p/977288#M21094</link>
      <description>Thanks.&lt;BR /&gt;I understood it.&lt;BR /&gt;Thanks.</description>
      <pubDate>Fri, 09 Sep 2005 07:51:56 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/even-mask-ippiFilterColumn-32f-C1R/m-p/977288#M21094</guid>
      <dc:creator>seiji-torigoe</dc:creator>
      <dc:date>2005-09-09T07:51:56Z</dc:date>
    </item>
  </channel>
</rss>

