opencv_traincascade (1)

opencv_traincascade [options]

traincascade application.

opencv_traincascade supports the following options:

-data cascade_dir_name

-vec vec_file_name

-bg background_file_name

-numPos number_of_positive_samples

The default is 2000.

-numNeg number_of_negative_samples

The default is 1000.

-num Stagesnumber_of_stages

The default is 20.

-precalcValBufSize precalculated_vals_buffer_size_in_Mb

The default is 256.

-precalcIdxBufSize precalculated_idxs_buffer_size_in_Mb

The default is 256.

-baseFormatSave

-stageType

The default is BOOST.

-featureType

Set feature type . You can select HAAR or LBP. The default is HAAR.

-w sampleWidth

The default is 24.

-h sampleHeight

The default is 24.

-bt {DAB|RAB|LB|GAB}

The type of the applied boosting algorithm. You can choose between Discrete AdaBoost (DAB), Real AdaBoost (RAB), LogitBoost (LB) and Gentle AdaBoost (GAB). The default is GAB.

-minHitRate min_hit_rate

The default is 0.995.

-maxFalseAlarmRate max_false_alarm_rate

The default is 0.5.

-weightTrimRate weight_trim_rate

The default is 0.95.

-maxDepth max_depth_of_weak_tree

The default is 1.

-maxWeakCount max_weak_tree_count

The default is 100.

-mode <BASIC|CORE|ALL>

The type of the applied haarFeature mode. You can choose between BASIC, ORE and ALL. The default is BASIC.

TODO

opencv_haartraing(1), opencv_performance(1)

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

This manual page was written by Nobuhiro Iwamatsu <[email protected]> for the Debian project (but may be used by others).