opencv_createsamples - create training and testing samples

       This   manual  page  was  written  by  DanielLeidert  <daniel.leidert@wgdd.de>  and  NobuhiroIwamatsu
       <iwamatsu@debian.org> for the Debian project (but may be used by others).

OpenCV                                              May 2010                             OPENCV_CREATESAMPLES(1)

opencv_createsamples  creates  positive  samples  from  a single object image or a collection of positive
       images.

       The scheme of test samples creation is similar to training samples creation since each test sample  is  a
       background  image  into  which a randomly distorted and randomly scaled instance of the object picture is
       pasted at a random position.

       To create training samples from one image applying distortions and show the results:

              opencv_createsamples-imgsource.png-num10-bgnegatives.dat-vecsamples_out.vec-show

       To create training samples of size 40x40 from some images without applying distortions:

              opencv_creasamples-infosource.dat-vecsamples_out.vec-w40-h40

       opencv_createsamples - create training and testing samples

opencv_createsamples supports the following options:

       -infocollection_file_name
              A database of positive source images. Use it together with -img to create test samples instead.

       -imgimage_file_name
              A positive source image. Use it together with -info to create test samples instead.

       -vecvec_file_name
              The output file name containing the generated positive samples for training.

       -bgbackground_file_name
              The background description file (the negative sample set). It contains a list of images into which
              randomly distorted versions of the object are pasted for positive sample generation.

       -numnumber_of_samples
              The number of positive samples to generate/train. The default is 1000.

       -bgcolorbackground_color
              The background color (currently grayscale images are assumed); the background  color  denotes  the
              transparent color. The default is 0.

       -inv   Invert the colors.

       -maxidevmax_intensity_deviation
              The desired maximum intensity deviation of foreground samples pixels. The default is 40.

       -maxxanglemax_x_rotation_angle
              The maximum rotation angle in x-direction in radians. The default is 1.1.

       -maxyanglemax_y_rotation_angle
              The maximum rotation angle in y-direction in radians. The default is 1.1.

       -maxzanglemax_z_rotation_angle
              The maximum rotation angle in z-direction in radians. The default is 0.5.

       -show[scale_factor]
              Show  each created sample during the creation process. Optionally a scaling factor can be defined.
              The default is 4.0.

              If <ESC> is pressed, the creation process will continue without showing the samples. This  can  be
              useful for debugging purposes.

       -hsample_height
              For the creation of training samples, it is the resulting sample height.  The default is 24.

              In case of creating test samples, it is the minimal height of placed object picture.

       -wsample_width
              For the creation of training samples, it is the resulting sample width.  The default is 24.

              In case of creating test samples, it is the minimal width of placed object picture.

       The same information is shown, if opencv_createsamples is called without any arguments/options.

opencv_haartraing(1), opencv_performance(1)

       More information and examples can be found in the OpenCV documentation.

opencv_createsamples[options]