<html> <head> <meta http-equiv=Content-Type content="text/html; charset=windows-1251"> <meta name=Generator content="Microsoft Word 11 (filtered)"> <title>Object Detection Using Haar-like Features with Cascade of Boosted Classifiers</title> <style> <!-- /* Style Definitions */ p.MsoNormal, li.MsoNormal, div.MsoNormal {margin:0in; margin-bottom:.0001pt; text-align:justify; font-size:12.0pt; font-family:"Times New Roman";} h1 {margin-top:12.0pt; margin-right:0in; margin-bottom:3.0pt; margin-left:0in; text-align:justify; page-break-after:avoid; font-size:16.0pt; font-family:Arial;} h2 {margin-top:12.0pt; margin-right:0in; margin-bottom:3.0pt; margin-left:0in; text-align:justify; page-break-after:avoid; font-size:14.0pt; font-family:Arial; font-style:italic;} h3 {margin-top:12.0pt; margin-right:0in; margin-bottom:3.0pt; margin-left:0in; text-align:justify; page-break-after:avoid; font-size:13.0pt; font-family:Arial;} span.Typewch {font-family:"Courier New"; font-weight:bold;} @page Section1 {size:595.3pt 841.9pt; margin:56.7pt 88.0pt 63.2pt 85.05pt;} div.Section1 {page:Section1;} /* List Definitions */ ol {margin-bottom:0in;} ul {margin-bottom:0in;} --> </style> </head> <body lang=RU> <div class=Section1> <h1><span lang=EN-US>Rapid Object Detection With A Cascade of Boosted Classifiers Based on Haar-like Features</span></h1> <h2><span lang=EN-US>Introduction</span></h2> <p class=MsoNormal><span lang=EN-US>This document describes how to train and use a cascade of boosted classifiers for rapid object detection. A large set of over-complete haar-like features provide the basis for the simple individual classifiers. Examples of object detection tasks are face, eye and nose detection, as well as logo detection. </span></p> <p class=MsoNormal><span lang=EN-US> </span></p> <p class=MsoNormal><span lang=EN-US>The sample detection task in this document is logo detection, since logo detection does not require the collection of large set of registered and carefully marked object samples. Instead we assume that from one prototype image, a very large set of derived object examples can be derived (</span><span class=Typewch><span lang=EN-US>createsamples</span></span><span lang=EN-US> utility, see below).</span></p> <p class=MsoNormal><span lang=EN-US> </span></p> <p class=MsoNormal><span lang=EN-US>A detailed description of the training/evaluation algorithm can be found in [1] and [2].</span></p> <h2><span lang=EN-US>Samples Creation</span></h2> <p class=MsoNormal><span lang=EN-US>For training a training samples must be collected. There are two sample types: negative samples and positive samples. Negative samples correspond to non-object images. Positive samples correspond to object images.</span></p> <h3><span lang=EN-US>Negative Samples</span></h3> <p class=MsoNormal><span lang=EN-US>Negative samples are taken from arbitrary images. These images must not contain object representations. Negative samples are passed through background description file. It is a text file in which each text line contains the filename (relative to the directory of the description file) of negative sample image. This file must be created manually. Note that the negative samples and sample images are also called background samples or background samples images, and are used interchangeably in this document</span></p> <p class=MsoNormal><span lang=EN-US> </span></p> <p class=MsoNormal><span lang=EN-US>Example of negative description file:</span></p> <p class=MsoNormal><span lang=EN-US> </span></p> <p class=MsoNormal><span lang=EN-US>Directory structure:</span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US>/img</span></span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US>� img1.jpg</span></span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US>� img2.jpg</span></span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US>bg.txt</span></span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US> </span></span></p> <p class=MsoNormal><span class=Typewch><span style='font-family:"Times New Roman"; font-weight:normal'>File </span></span><span class=Typewch><span lang=EN-US>bg.txt:</span></span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US>img/img1.jpg</span></span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US>img/img2.jpg</span></span></p> <h3><span lang=EN-US>Positive Samples</span></h3> <p class=MsoNormal><span lang=EN-US>Positive samples are created by </span><span class=Typewch><span lang=EN-US>createsamples</span></span><span lang=EN-US> utility. They may be created from single object image or from collection of previously marked up images.<br> <br> </span></p> <p class=MsoNormal><span lang=EN-US>The single object image may for instance contain a company logo. Then are large set of positive samples are created from the given object image by randomly rotating, changing the logo color as well as placing the logo on arbitrary background.</span></p> <p class=MsoNormal><span lang=EN-US>The amount and range of randomness can be controlled by command line arguments. </span></p> <p class=MsoNormal><span lang=EN-US>Command line arguments:</span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- vec <vec_file_name></span></span><span lang=EN-US> </span></p> <p class=MsoNormal style='margin-left:17.1pt'><span lang=EN-US>name of the output file containing the positive samples for training</span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- img <image_file_name></span></span><span lang=EN-US> </span></p> <p class=MsoNormal style='margin-left:17.1pt'><span lang=EN-US>source object image (e.g., a company logo)</span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- bg <background_file_name></span></span><span lang=EN-US> </span></p> <p class=MsoNormal style='margin-left:17.1pt'><span lang=EN-US>background description file; contains a list of images into which randomly distorted versions of the object are pasted for positive sample generation</span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- num <number_of_samples></span></span><span lang=EN-US> </span></p> <p class=MsoNormal style='margin-left:17.1pt'><span lang=EN-US>number of positive samples to generate </span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- bgcolor <background_color></span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span lang=EN-US>����� background color (currently grayscale images are assumed); the background color denotes the transparent color. Since there might be compression artifacts, the amount of color tolerance can be specified by </span><span class=Typewch><span lang=EN-US>�bgthresh</span></span><span class=Typewch><span lang=EN-US style='font-family:Arial;font-weight:normal'>. </span></span><span lang=EN-US>All pixels between </span><span class=Typewch><span lang=EN-US>bgcolor-bgthresh</span></span><span lang=EN-US> and </span><span class=Typewch><span lang=EN-US>bgcolor+bgthresh</span></span><span lang=EN-US> are regarded as transparent.</span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- bgthresh <background_color_threshold></span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- inv</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span lang=EN-US>����� if specified, the colors will be inverted</span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- randinv</span></span><span lang=EN-US> </span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span lang=EN-US>����� if specified, the colors will be inverted randomly</span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- maxidev <max_intensity_deviation></span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>� </span></span><span lang=EN-US>maximal intensity deviation of foreground samples pixels</span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- maxxangle <max_x_rotation_angle>,</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- maxyangle <max_y_rotation_angle>,</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- maxzangle <max_z_rotation_angle></span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span lang=EN-US>����� maximum rotation angles in radians</span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>-show</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span lang=EN-US>����� if specified, each sample will be shown. Pressing �Esc� will continue creation process without samples showing. Useful debugging option.</span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- w <sample_width></span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>� </span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight:normal'>width (in pixels) of the output samples</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- h <sample_height></span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>� </span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight:normal'>height (in pixels) of the output samples</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US> </span></span></p> <p class=MsoNormal><span lang=EN-US>For following procedure is used to create a sample object instance:</span></p> <p class=MsoNormal><span lang=EN-US>The source image is rotated random around all three axes. The chosen angle is limited my</span><span class=Typewch><span lang=EN-US> -max?angle</span></span><span lang=EN-US>. Next pixels of intensities in the range of </span><span class=Typewch><span lang=EN-US>[bg_color-bg_color_threshold; bg_color+bg_color_threshold]</span></span><span lang=EN-US> are regarded as transparent. White noise is added to the intensities of the foreground. If </span><span class=Typewch><span lang=EN-US>�inv</span></span><span lang=EN-US> key is specified then foreground pixel intensities are inverted. If </span><span class=Typewch><span lang=EN-US>�randinv</span></span><span lang=EN-US> key is specified then it is randomly selected whether for this sample inversion will be applied. Finally, the obtained image is placed onto arbitrary background from the background description file, resized to the pixel size specified by </span><span class=Typewch><span lang=EN-US>�w</span></span><span lang=EN-US> and </span><span class=Typewch><span lang=EN-US>�h</span></span><span lang=EN-US> and stored into the file specified by the </span><span class=Typewch><span lang=EN-US>�vec</span></span><span lang=EN-US> command line parameter.</span></p> <p class=MsoNormal><span lang=EN-US> </span></p> <p class=MsoNormal><span lang=EN-US>Positive samples also may be obtained from a collection of previously marked up images. This collection is described by text file similar to background description file. Each line of this file corresponds to collection image. The first element of the line is image file name. It is followed by number of object instances. The following numbers are the coordinates of bounding rectangles (x, y, width, height).</span></p> <p class=MsoNormal><span lang=EN-US> </span></p> <p class=MsoNormal><span lang=EN-US>Example of description file:</span></p> <p class=MsoNormal><span lang=EN-US> </span></p> <p class=MsoNormal><span lang=EN-US>Directory structure:</span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US>/img</span></span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US>� img1.jpg</span></span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US>� img2.jpg</span></span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US>info.dat</span></span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US> </span></span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US style='font-family: "Times New Roman";font-weight:normal'>File </span></span><span class=Typewch><span lang=EN-US>info.dat:</span></span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US>img/img1.jpg� 1� 140 100 45 45</span></span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US>img/img2.jpg� 2� 100 200 50 50�� 50 30 25 25</span></span></p> <p class=MsoNormal><span lang=EN-US> </span></p> <p class=MsoNormal><span lang=EN-US>Image </span><span class=Typewch><span lang=EN-US>img1.jpg</span></span><span lang=EN-US> contains single object instance with bounding rectangle (140, 100, 45, 45). Image </span><span class=Typewch><span lang=EN-US>img2.jpg</span></span><span lang=EN-US> contains two object instances.</span></p> <p class=MsoNormal><span lang=EN-US> </span></p> <p class=MsoNormal><span lang=EN-US>In order to create positive samples from such collection </span><span class=Typewch><span lang=EN-US>�info</span></span><span lang=EN-US> argument should be specified instead of </span><span class=Typewch><span lang=EN-US>�img</span></span><span class=Typewch><span style='font-family:"Times New Roman"; font-weight:normal'>:</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- info <collection_file_name></span></span><span lang=EN-US> </span></p> <p class=MsoNormal style='margin-left:17.1pt'><span lang=EN-US>description file of marked up images collection</span></p> <p class=MsoNormal><span lang=EN-US> </span></p> <p class=MsoNormal><span lang=EN-US>The scheme of sample creation in this case is as follows. The object instances are taken from images. Then they are resized to samples size and stored in output file. No distortion is applied, so the only affecting arguments are </span><span class=Typewch><span lang=EN-US>�w</span></span><span lang=EN-US>, </span><span class=Typewch><span lang=EN-US>-h</span></span><span lang=EN-US>, </span><span class=Typewch><span lang=EN-US>-show</span></span><span lang=EN-US> and </span><span class=Typewch><span lang=EN-US>�num</span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>.</span></span></p> <p class=MsoNormal><span lang=EN-US> </span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US>createsamples</span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'> utility may be used for examining samples stored in positive samples file. In order to do this only </span></span><span class=Typewch><span lang=EN-US>�vec</span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight:normal'>, </span></span><span class=Typewch><span lang=EN-US>�w</span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight:normal'> and </span></span><span class=Typewch><span lang=EN-US>�h</span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight:normal'> parameters should be specified.</span></span></p> <p class=MsoNormal><span lang=EN-US> </span></p> <p class=MsoNormal><span lang=EN-US>Note that for training, it does not matter how positive samples files are generated. So the </span><span class=Typewch><span lang=EN-US>createsamples</span></span><span lang=EN-US> utility is only one way to collect/create a vector file of positive samples.</span></p> <h2><span lang=EN-US>Training</span></h2> <p class=MsoNormal><span lang=EN-US>The next step after samples creation is training of classifier. It is performed by the </span><span class=Typewch><span lang=EN-US>haartraining</span></span><span lang=EN-US> utility.</span></p> <p class=MsoNormal><span lang=EN-US> </span></p> <p class=MsoNormal><span lang=EN-US>Command line arguments:</span><span class=Typewch><span lang=EN-US> </span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- data <dir_name></span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'> </span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>����� directory name in which the trained classifier is stored</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- vec <vec_file_name></span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'> </span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>����� file name of positive sample file (created by </span></span><span class=Typewch><span lang=EN-US>trainingsamples</span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'> utility or by any other means)</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- bg <background_file_name></span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>����� background description file</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- npos <number_of_positive_samples>,</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- nneg <number_of_negative_samples></span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'> </span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>����� number of positive/negative samples used in training of each classifier stage. Reasonable values are npos = 7000 and nneg = 3000.</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- nstages <number_of_stages></span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>� </span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight:normal'>number of stages to be trained</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- nsplits <number_of_splits></span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'> </span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>����� determines the weak classifier used in stage classifiers. If </span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman"'>1</span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>, then a simple stump classifier is used, if </span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman"'>2</span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'> and more, then CART classifier with </span></span><span class=Typewch><span lang=EN-US>number_of_splits</span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight:normal'> internal (split) nodes is used</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- mem <memory_in_MB></span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'> </span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>����� Available memory in MB for precalculation. The more memory you have the faster the training process</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- sym (default),</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- nonsym</span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight:normal'> </span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>����� specifies whether the object class under training has vertical symmetry or not. Vertical symmetry speeds up training process. For instance, frontal faces show off vertical symmetry</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- minhitrate <min_hit_rate></span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>����� minimal desired hit rate for each stage classifier. Overall hit rate may be estimated as </span></span><span class=Typewch><span lang=EN-US>(min_hit_rate^number_of_stages)</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- maxfalsealarm <max_false_alarm_rate></span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'> </span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>����� maximal desired false alarm rate for each stage classifier. </span></span><span class=Typewch><span style='font-family:"Times New Roman";font-weight:normal'>Overall false alarm rate may be estimated as</span></span><span class=Typewch><span lang=EN-US> (max_false_alarm_rate^number_of_stages)</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- weighttrimming <weight_trimming></span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>� </span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight:normal'>Specifies whether and how much weight trimming should be used. A decent choice is 0.90.</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- eqw</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- mode <BASIC (default) | CORE | ALL></span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'> </span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>����� selects the type of haar features set used in training. BASIC use only upright features, while ALL uses the full set of upright and 45 degree rotated feature set. See [1] for more details.</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- w <sample_width>,</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- h <sample_height></span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'> </span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>����� Size of training samples (in pixels). Must have exactly the same values as used during training samples creation (utility </span></span><span class=Typewch><span lang=EN-US>trainingsamples</span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>)</span></span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US style='font-family: "Times New Roman";font-weight:normal'> </span></span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US style='font-family: "Times New Roman";font-weight:normal'>Note: in order to use multiprocessor advantage a compiler that supports OpenMP 1.0 standard should be used.</span></span></p> <h2><span lang=EN-US>Application</span></h2> <p class=MsoNormal><span lang=EN-US>OpenCV cvHaarDetectObjects() function (in particular haarFaceDetect demo) is used for detection.</span></p> <h3><span lang=EN-US>Test Samples</span></h3> <p class=MsoNormal><span lang=EN-US>In order to evaluate the performance of trained classifier a collection of marked up images is needed. When such collection is not available test samples may be created from single object image by </span><span class=Typewch><span lang=EN-US>createsamples</span></span><span lang=EN-US> utility. The scheme of test samples creation in this case 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. </span></p> <p class=MsoNormal><span lang=EN-US> </span></p> <p class=MsoNormal><span lang=EN-US>If both </span><span class=Typewch><span lang=EN-US>�img</span></span><span lang=EN-US> and </span><span class=Typewch><span lang=EN-US>�info</span></span><span lang=EN-US> arguments are specified then test samples will be created by </span><span class=Typewch><span lang=EN-US>createsamples</span></span><span lang=EN-US> utility. The sample image is arbitrary distorted as it was described below, then it is placed at random location to background image and stored. The corresponding description line is added to the file specified by </span><span class=Typewch><span lang=EN-US>�info</span></span><span lang=EN-US> argument.</span></p> <p class=MsoNormal><span lang=EN-US> </span></p> <p class=MsoNormal><span lang=EN-US>The </span><span class=Typewch><span lang=EN-US>�w</span></span><span lang=EN-US> and </span><span class=Typewch><span lang=EN-US>�h</span></span><span lang=EN-US> keys determine the minimal size of placed object picture.</span></p> <p class=MsoNormal><span lang=EN-US> </span></p> <p class=MsoNormal><span lang=EN-US>The test image file name format is as follows:</span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US>imageOrderNumber_x_y_width_height.jpg</span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>, where </span></span><span class=Typewch><span lang=EN-US>x</span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>, </span></span><span class=Typewch><span lang=EN-US>y</span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>, </span></span><span class=Typewch><span lang=EN-US>width</span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'> and </span></span><span class=Typewch><span lang=EN-US>height</span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'> are the coordinates of placed object bounding rectangle.</span></span></p> <p class=MsoNormal><span class=Typewch><span lang=EN-US style='font-family: "Times New Roman";font-weight:normal'>Note that you should use a background images set different from the background image set used during training.</span></span></p> <h3><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman"'>Performance Evaluation</span></span></h3> <p class=MsoNormal><span lang=EN-US>In order to evaluate the performance of the classifier </span><span class=Typewch><span lang=EN-US>performance</span></span><span lang=EN-US> utility may be used. It takes a collection of marked up images, applies the classifier and outputs the performance, i.e. number of found objects, number of missed objects, number of false alarms and other information.</span></p> <p class=MsoNormal><span lang=EN-US> </span></p> <p class=MsoNormal><span lang=EN-US>Command line arguments:</span><span class=Typewch><span lang=EN-US> </span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- data <dir_name></span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'> </span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>����� directory name in which the trained classifier is stored</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- info <collection_file_name></span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'> </span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>����� file with test samples description</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- maxSizeDiff <max_size_difference></span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>,</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- maxPosDiff <max_position_difference></span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'> </span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>����� determine the criterion of reference and detected rectangles coincidence. Default values are 1.5 and 0.3 respectively.</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- sf <scale_factor></span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>,</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>����� detection parameter. Default value is 1.2.</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- w <sample_width>,</span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US>- h <sample_height></span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'> </span></span></p> <p class=MsoNormal style='margin-left:17.1pt;text-indent:-17.1pt'><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight: normal'>����� Size of training samples (in pixels). Must have exactly the same values as used during training (utility </span></span><span class=Typewch><span lang=EN-US>haartraining</span></span><span class=Typewch><span lang=EN-US style='font-family:"Times New Roman";font-weight:normal'>)</span></span></p> <h2><span lang=EN-US>References</span></h2> <p class=MsoNormal><span lang=EN-US>[1] Rainer Lienhart and Jochen Maydt. An Extended Set of Haar-like Features for Rapid Object Detection. Submitted to ICIP2002.</span></p> <p class=MsoNormal><span lang=EN-US>[2] Alexander Kuranov, Rainer Lienhart, and Vadim Pisarevsky. An Empirical Analysis of Boosting Algorithms for Rapid Objects With an Extended Set of Haar-like Features. Intel Technical Report MRL-TR-July02-01, 2002.</span></p> </div> </body> </html>