1 .TH "OPENCV\-PERFORMANCE" "1" "May 2008" "OpenCV" "User Commands"
5 opencv-performance \- evaluate the performance of the classifier
9 .B opencv\-performance [options]
14 .B opencv\-performance
15 evaluates the performance of the classifier. It takes a collection of marked
16 up test images, applies the classifier and outputs the performance, i.e. number of
17 found objects, number of missed objects, number of false alarms and other
20 When there is no such collection available test samples may be created from single
22 .BR opencv\-createsamples (1)
23 utility. The scheme of test samples creation in this case is similar to training samples
25 In the output, the table should be read:
28 shows the number of correctly found objects
31 shows the number of missed objects (must exist but are not found, also known
35 shows the number of false alarms (must not exist but are found, also known
41 .B opencv\-performance
42 supports the following options:
47 .BI "\-data " classifier_directory_name
48 The directory, in which the classifier can be found.
51 .BI "\-info " collection_file_name
52 File with test samples description.
55 .BI "\-maxSizeDiff " max_size_difference
56 Determine the size criterion of reference and detected coincidence.
61 .BI "\-maxPosDiff " max_position_difference
62 Determine the position criterion of reference and detected coincidence.
67 .BI "\-sf " scale_factor
68 Scale the detection window in each iteration. The default is
73 Don't save detection result to an image. This could be useful, if
74 .I collection_file_name
78 .BI "\-nos " number_of_stages
79 Number of stages to use. The default is
81 (all stages are used).
89 .BI "\-h " sample_height
90 The sample height (must have the same value as used during creation).
95 .BI "\-w " sample_width
96 The sample width (must have the same value as used during creation).
101 The same information is shown, if
102 .B opencv\-performance
103 is called without any arguments/options.
108 To create training samples from one image applying distortions and show the
111 .B opencv\-performance -data trainout -info tests.dat
116 .BR opencv\-createsamples (1),
117 .BR opencv\-haartraing (1)
119 More information and examples can be found in the OpenCV documentation.
124 This manual page was written by \fBDaniel Leidert\fR <\&daniel.leidert@wgdd.de\&>
125 for the Debian project (but may be used by others).