<?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 Re: IPPI Morphological operations vs. Min/Max Filtering in Intel® Integrated Performance Primitives</title>
    <link>https://community.intel.com/t5/Intel-Integrated-Performance/IPPI-Morphological-operations-vs-Min-Max-Filtering/m-p/985705#M21782</link>
    <description>&lt;DIV&gt;Hi Adi,&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;you can compare performance of different IPP functions on different architectures, just find IPP performance data in your IPP	oolsperfsysdata folder.&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;For example, ippiFilterMax_8u_C1 function shows the folloving performance on 3050MHz Pentium 4 with HT tech with image 720x480:&lt;/DIV&gt;
&lt;DIV&gt;ippiFilterMax_8u_C1R, image 720x480, kernel 3x3,4.3 cpe, 496 usec&lt;BR /&gt;ippiFilterMax_8u_C1R, image 720x480, kernel 5x5,6.4 cpe, 723 usec&lt;BR /&gt;&lt;/DIV&gt;
&lt;DIV&gt;whereas performance ofippiDilate_8u_C1 function in the same conditionslooks like:&lt;/DIV&gt;
&lt;DIV&gt;ippiDilate_8u_C1R, image 720x480, kernel 3x3, 8.1 cpe, 922 usec&lt;BR /&gt;ippiDilate_8u_C1R, image 720x480, kernel 5x5, 20 cpe, 2.35e+003 usec&lt;BR /&gt;&lt;/DIV&gt;
&lt;DIV&gt;Here,&lt;/DIV&gt;
&lt;DIV&gt;cpe is cpu clock per pixel
&lt;DIV&gt;usec is microseconds per image&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;Regards,&lt;/DIV&gt;
&lt;DIV&gt; Vladimir&lt;/DIV&gt;</description>
    <pubDate>Fri, 12 Aug 2005 22:52:58 GMT</pubDate>
    <dc:creator>Vladimir_Dudnik</dc:creator>
    <dc:date>2005-08-12T22:52:58Z</dc:date>
    <item>
      <title>IPPI Morphological operations vs. Min/Max Filtering</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/IPPI-Morphological-operations-vs-Min-Max-Filtering/m-p/985702#M21779</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;What is the difference between the "morphological operations" e.g. ippiDilate_() and the filtering functions e.g. ippiFilterMax_(). These are defined to be identical.&lt;BR /&gt;&lt;BR /&gt;Is there an implementation difference?&lt;BR /&gt;Is there a performance difference?&lt;BR /&gt;And if so why?&lt;BR /&gt;&lt;BR /&gt;Thanks,&lt;BR /&gt;Adi</description>
      <pubDate>Tue, 12 Jul 2005 19:57:10 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/IPPI-Morphological-operations-vs-Min-Max-Filtering/m-p/985702#M21779</guid>
      <dc:creator>adi_shavit</dc:creator>
      <dc:date>2005-07-12T19:57:10Z</dc:date>
    </item>
    <item>
      <title>Re: IPPI Morphological operations vs. Min/Max Filtering</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/IPPI-Morphological-operations-vs-Min-Max-Filtering/m-p/985703#M21780</link>
      <description>&lt;DIV&gt;Hi Adi,&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;&lt;FONT color="#000080" size="2"&gt;
&lt;P&gt;Min/Max filters work as general filters  with rectangular kernel. Morphological functions with rectangular kernel may have the same realization as Min/Max filters, but in addition they support arbitrary kernel/mask. So they have different API and functionality.&lt;/P&gt;
&lt;P&gt;Regards,&lt;BR /&gt; Vladimir&lt;/P&gt;&lt;/FONT&gt;&lt;/DIV&gt;</description>
      <pubDate>Wed, 27 Jul 2005 22:54:12 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/IPPI-Morphological-operations-vs-Min-Max-Filtering/m-p/985703#M21780</guid>
      <dc:creator>Vladimir_Dudnik</dc:creator>
      <dc:date>2005-07-27T22:54:12Z</dc:date>
    </item>
    <item>
      <title>Re: IPPI Morphological operations vs. Min/Max Filtering</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/IPPI-Morphological-operations-vs-Min-Max-Filtering/m-p/985704#M21781</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;  I see, so basically for rectangular kernels the performance should be the same, or is the "simpler" implementation of min/max faster?&lt;BR /&gt;&lt;BR /&gt;(Of course, I can check but I'm wondering implementation-wise.)&lt;BR /&gt;&lt;BR /&gt;Adi</description>
      <pubDate>Thu, 28 Jul 2005 02:40:21 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/IPPI-Morphological-operations-vs-Min-Max-Filtering/m-p/985704#M21781</guid>
      <dc:creator>adi_shavit</dc:creator>
      <dc:date>2005-07-28T02:40:21Z</dc:date>
    </item>
    <item>
      <title>Re: IPPI Morphological operations vs. Min/Max Filtering</title>
      <link>https://community.intel.com/t5/Intel-Integrated-Performance/IPPI-Morphological-operations-vs-Min-Max-Filtering/m-p/985705#M21782</link>
      <description>&lt;DIV&gt;Hi Adi,&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;you can compare performance of different IPP functions on different architectures, just find IPP performance data in your IPP	oolsperfsysdata folder.&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;For example, ippiFilterMax_8u_C1 function shows the folloving performance on 3050MHz Pentium 4 with HT tech with image 720x480:&lt;/DIV&gt;
&lt;DIV&gt;ippiFilterMax_8u_C1R, image 720x480, kernel 3x3,4.3 cpe, 496 usec&lt;BR /&gt;ippiFilterMax_8u_C1R, image 720x480, kernel 5x5,6.4 cpe, 723 usec&lt;BR /&gt;&lt;/DIV&gt;
&lt;DIV&gt;whereas performance ofippiDilate_8u_C1 function in the same conditionslooks like:&lt;/DIV&gt;
&lt;DIV&gt;ippiDilate_8u_C1R, image 720x480, kernel 3x3, 8.1 cpe, 922 usec&lt;BR /&gt;ippiDilate_8u_C1R, image 720x480, kernel 5x5, 20 cpe, 2.35e+003 usec&lt;BR /&gt;&lt;/DIV&gt;
&lt;DIV&gt;Here,&lt;/DIV&gt;
&lt;DIV&gt;cpe is cpu clock per pixel
&lt;DIV&gt;usec is microseconds per image&lt;/DIV&gt;
&lt;DIV&gt;&lt;/DIV&gt;&lt;/DIV&gt;
&lt;DIV&gt;Regards,&lt;/DIV&gt;
&lt;DIV&gt; Vladimir&lt;/DIV&gt;</description>
      <pubDate>Fri, 12 Aug 2005 22:52:58 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Integrated-Performance/IPPI-Morphological-operations-vs-Min-Max-Filtering/m-p/985705#M21782</guid>
      <dc:creator>Vladimir_Dudnik</dc:creator>
      <dc:date>2005-08-12T22:52:58Z</dc:date>
    </item>
  </channel>
</rss>

