Paparazzi UAS  v5.10_stable-5-g83a0da5-dirty
Paparazzi is a free software Unmanned Aircraft System.
 All Data Structures Namespaces Files Functions Variables Typedefs Enumerations Enumerator Macros Modules Pages
cv_textons module

Represent the appearance (texture, color) of an image by means of a texton histogram.

No detailed description...

Example for airframe file

Add to your firmware section:

1 <module name="cv_textons"/>

Module configuration options

Airframe file section

  • section name: TEXTONS prefix: TEXTONS_
    • name LOAD_DICTIONARY value: YES
      Description: Whether a dictionary is loaded (YES) or learned (NO).
    • name N_TEXTONS value: 20
      Description: The number of textons (words) in the dictionary.
    • name PATCH_SIZE value: 6
      Description: Size of the image patches extracted from the image - even numbers.
    • name FULL_SAMPLING value: NO
      Description: If YES, the entire image is covered with samples, if NO sub-sampling is used (faster).
    • name DICTIONARY_NUMBER value: 0
      Description: Number of the dictionary (so that different environments can each have their specific dictionary).
    • name ALPHA value: 10
      Description: Learning rate when creating the dictionary: 0 = no learning, 255 = the texton becomes equal to the new patch.
    • name N_SAMPLES_IMAGE value: 50
      Description: Number of samples extracted form the image when not doing full sampling.
    • name N_LEARNING_SAMPLES value: 10000
      Description: Number of samples used for learning the dictionary.
    • name BORDER_WIDTH value: 0
      Description: Width of the image border from which no samples are taken.
    • name BORDER_HEIGHT value: 0
      Description: Height of the border from which no samples are taken.

Module functions

Init Functions

These initialization functions are called once on startup.

Files

Header Files

The following headers are automatically included in modules.h

Source Files

Raw cv_textons.xml file:

<!DOCTYPE module SYSTEM "module.dtd">
<module name="cv_textons" dir="computer_vision">
<doc>
<description>Represent the appearance (texture, color) of an image by means of a texton histogram.</description>
<section name="TEXTONS" prefix="TEXTONS_">
<define name="LOAD_DICTIONARY" value="YES" description="Whether a dictionary is loaded (YES) or learned (NO)."/>
<define name="N_TEXTONS" value="20" description="The number of textons (words) in the dictionary."/>
<define name="PATCH_SIZE" value="6" description="Size of the image patches extracted from the image - even numbers."/>
<define name="FULL_SAMPLING" value="NO" description="If YES, the entire image is covered with samples, if NO sub-sampling is used (faster)."/>
<define name="DICTIONARY_NUMBER" value="0" description="Number of the dictionary (so that different environments can each have their specific dictionary)."/>
<define name="ALPHA" value="10" description="Learning rate when creating the dictionary: 0 = no learning, 255 = the texton becomes equal to the new patch."/>
<define name="N_SAMPLES_IMAGE" value="50" description="Number of samples extracted form the image when not doing full sampling."/>
<define name="N_LEARNING_SAMPLES" value="10000" description="Number of samples used for learning the dictionary."/>
<define name="BORDER_WIDTH" value="0" description="Width of the image border from which no samples are taken."/>
<define name="BORDER_HEIGHT" value="0" description="Height of the border from which no samples are taken."/>
</section>
</doc>
<settings>
<dl_settings>
<dl_settings NAME="Textons">
<dl_setting var="load_dictionary" min="0" step="1" max="1" module="computer_vision/textons" shortname="load dict." values="NO|YES" param="TEXTONS_LOAD_DICTIONARY"/>
<dl_setting var="n_textons" min="1" step="1" max="255" shortname="n_textons" param="TEXTONS_N_TEXTONS"/>
<dl_setting var="patch_size" min="2" step="2" max="40" shortname="p_size" param="TEXTONS_PATCH_SIZE" />
<dl_setting var="FULL_SAMPLING" min="0" step="1" max="1" shortname="full_sam" values="NO|YES" param="TEXTONS_FULL_SAMPLING" />
<dl_setting var="dictionary_number" min="0" step="1" max="20" shortname="dict_num" param="TEXTONS_DICTIONARY_NUMBER" />
<dl_setting var="n_samples_image" min="0" step="10" max="1000" shortname="n_samples" param="TEXTONS_ALPHA" />
<dl_setting var="n_learning_samples" min="1" step="100" max="250000" shortname="n_l_samples" param="TEXTONS_N_LEARNING_SAMPLES" />
<dl_setting var="alpha" min="0" step="1" max="255" shortname="alpha" param="TEXTONS_ALPHA"/>
<dl_setting var="border_width" min="0" step="1" max="200" shortname="b_w" param="TEXTONS_BORDER_WIDTH"/>
<dl_setting var="border_height" min="0" step="1" max="200" shortname="b_h" param="TEXTONS_BORDER_HEIGHT"/>
</dl_settings>
</dl_settings>
</settings>
<header>
<file name="textons.h"/>
</header>
<init fun="textons_init()"/>
<makefile target="ap">
<file name="textons.c"/>
<file name="image.c" dir="modules/computer_vision/lib/vision"/>
</makefile>
</module>