ImageMiner 1.3.3 - help documentation?

Doloop

Well, I finally managed to install the ImageMiner extension on a Windows XP system (never got it to work on my Mac OSX system).  I got it from here: [(http://spl.utko.feec.vutbr.cz/en/component/content/article/46-image-processing-extension-for-rapidminer-5)]  Now, I am trying to use it but it is not at all intuitive.  I cannot find any help documentation on their website.  Can anyone point me to some help documentation on the ImageMiner extension?

StaryVena

Hi,
on my http://www.myexperiment.org/users/18932/workflows are some workflows. You can start here or write me, what would you like to do. So I can post here some temp process based on your demands.

Best
Vaclav

Doloop

Thanks for the reply.  I will check out your workflows.  By the way, I finally got it working on my "big computer" (my Mac) so I am smiling now.  (I made another post for other users who may be having the same installation problems I was having.)

I want to classify images.  I have images from security cameras and I want to classify them based on whether they contain people or vehicles or do not contain people or vehicles.  The cameras are stationary.  I have many examples of images in which people and / or vehicles do appear and i also have many images in which they do not appear.  I also have many images in different levels of light (bright, sunny day, overcast day, etc).  I also have images of days in which there was a significant change in the landscape appearance (i.e., snowy days).  It should be fairly straight forward to implement especially if i break the recognition into different tasks.

  It looks like ImageMiner has the capability to do what I want to do.  I just need to figure out what all of the ImageMiner processes do so I know which ones would be helpful for my task.  I plan to experiment with both neural networks and also SVM's to perform the classification / recognition task.

StaryVena

Hi,
there is small example, how to convert images in tree folders to labeled example set:

<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<process version="5.1.013">
  <context>
    <input/>
    <output/>
    <macros/>
  </context>
  <operator activated="true" class="process" compatibility="5.1.013" expanded="true" name="Process">
    <process expanded="true" height="190" width="413">
      <operator activated="true" class="imageprocessing:multiple_color_image_opener" compatibility="1.2.000" expanded="true" height="60" name="MCIO" width="90" x="313" y="30">
        <list key="images">
          <parameter key="empty" value="dirWithoutObjet"/>
          <parameter key="cars" value="dirWithPeople"/>
          <parameter key="people" value="dirWithCars"/>
        </list>
        <parameter key="assign_label" value="true"/>
        <process expanded="true" height="553" width="1064">
          <operator activated="true" class="imageprocessing:global_feature_extraction" compatibility="1.2.000" expanded="true" height="60" name="Global Feature Extractor from a Single Image" width="90" x="380" y="30">
            <process expanded="true" height="553" width="1064">
              <operator activated="true" class="imageprocessing:statistics" compatibility="1.2.000" expanded="true" height="60" name="Global statistics" width="90" x="447" y="30"/>
              <operator activated="true" class="imageprocessing:histogram" compatibility="1.2.000" expanded="true" height="60" name="histogram" width="90" x="447" y="120"/>
              <operator activated="true" class="imageprocessing:color_to_grayscale" compatibility="1.2.000" expanded="true" height="60" name="Color to grayscale" width="90" x="179" y="210"/>
              <operator activated="true" class="imageprocessing:dLog_distance" compatibility="1.2.000" expanded="true" height="60" name="dLog" width="90" x="447" y="210"/>
              <connect from_port="color image plus 1" to_op="Global statistics" to_port="color image plus"/>
              <connect from_port="color image plus 2" to_op="histogram" to_port="color image plus"/>
              <connect from_port="color image plus 3" to_op="Color to grayscale" to_port="color image plus"/>
              <connect from_op="Global statistics" from_port="features" to_port="feature 1"/>
              <connect from_op="histogram" from_port="features" to_port="feature 2"/>
              <connect from_op="Color to grayscale" from_port="grayscale image" to_op="dLog" to_port="grayscale image plus Hist"/>
              <connect from_op="dLog" from_port="features" to_port="feature 3"/>
              <portSpacing port="source_color image plus 1" spacing="0"/>
              <portSpacing port="source_color image plus 2" spacing="72"/>
              <portSpacing port="source_color image plus 3" spacing="0"/>
              <portSpacing port="source_color image plus 4" spacing="0"/>
              <portSpacing port="sink_feature 1" spacing="0"/>
              <portSpacing port="sink_feature 2" spacing="72"/>
              <portSpacing port="sink_feature 3" spacing="0"/>
              <portSpacing port="sink_feature 4" spacing="0"/>
            </process>
          </operator>
          <connect from_port="color image plus" to_op="Global Feature Extractor from a Single Image" to_port="color image plus"/>
          <connect from_op="Global Feature Extractor from a Single Image" from_port="example set" to_port="Example set"/>
          <portSpacing port="source_color image plus" spacing="0"/>
          <portSpacing port="sink_Example set" spacing="0"/>
        </process>
      </operator>
      <connect from_op="MCIO" from_port="example set" to_port="result 1"/>
      <portSpacing port="source_input 1" spacing="0"/>
      <portSpacing port="sink_result 1" spacing="0"/>
      <portSpacing port="sink_result 2" spacing="0"/>
    </process>
  </operator>
</process>

Maybe some preprocessing operators are needed for improve performance (equalize histogram etc.)

Best
Vaclav

Doloop

Vaclav, thanks for the help.  I now have both a neural network and an svm evaluating my images.  I will have to work with the models to optimize and see which works best. 

I still would like to see some documentation on the ImageMiner extension.  Surely there is some somewhere.  Do you know where I might find it?

By the way, when I looked at MyExperiment to see your workflows, I read about Taverna.  I had never heard of that application before.  Are you familiar with it?  I am curious as to how it is being used.

Thanks!

StaryVena

Doloop wrote:
I still would like to see some documentation on the ImageMiner extension.  Surely there is some somewhere.  Do you know where I might find it?

Documentation is a big problem. Some preprocessing operators are documented on wiki http://rapid-i.com/wiki/index.php?title=Category:Image_extension
Situation about documentation will get better this summer. I hope 

Doloop wrote:
By the way, when I looked at MyExperiment to see your workflows, I read about Taverna.  I had never heard of that application before.  Are you familiar with it?  I am curious as to how it is being used.

I also had never heard of Taverna

Best
Vaclav

IngoRM

Hi,

Taverna is an open source workflow system for the execution of scientific workflows composed of different web services:

http://www.taverna.org.uk/

It has been mainly developed by the University of Manchester, one of Rapid-I's project partners in the European project "e-LICO". Taverna is a great system - especially in the domain of bioinformatics. And Taverna now offers a RapidMiner extension which can be used to address all RapidMiner operators as web services provided by a RapidAnalytics server - and this makes Taverna an even greater piece of software now 

More information about this extension can be found at

http://www.e-lico.eu/?q=TavernaRM

The extension was also presented during RCOMM 2011, our last user conference. The proceedings are available at

http://www.amazon.de/Proceedings-RapidMiner-Community-Conference-Informatik/dp/3844000933

Cheers,
Ingo

Doloop

Vaclav, thanks for the link to the wiki.  I've checked it out and do find it helpful, at least as a brief introduction to some of the operators.

Ingo, thanks for the info.  I will look into Taverna.  At first glance, it appears that it could be useful for a very wide range of tasks.