reduce_ocr_class_svmT_reduce_ocr_class_svmReduceOcrClassSvmReduceOcrClassSvmreduce_ocr_class_svm (Operator)
Name
reduce_ocr_class_svmT_reduce_ocr_class_svmReduceOcrClassSvmReduceOcrClassSvmreduce_ocr_class_svm — Approximate a trained SVM-based OCR classifier by a reduced SVM.
Signature
void ReduceOcrClassSvm(const HTuple& OCRHandle, const HTuple& Method, const HTuple& MinRemainingSV, const HTuple& MaxError, HTuple* OCRHandleReduced)
HOCRSvm HOCRSvm::ReduceOcrClassSvm(const HString& Method, Hlong MinRemainingSV, double MaxError) const
HOCRSvm HOCRSvm::ReduceOcrClassSvm(const char* Method, Hlong MinRemainingSV, double MaxError) const
HOCRSvm HOCRSvm::ReduceOcrClassSvm(const wchar_t* Method, Hlong MinRemainingSV, double MaxError) const
(
Windows only)
Description
reduce_ocr_class_svmreduce_ocr_class_svmReduceOcrClassSvmReduceOcrClassSvmreduce_ocr_class_svm reduces the classification time of an
SVM based OCR classifier OCRHandleOCRHandleOCRHandleOCRHandleocrhandle by returning a reduced
copy of it in OCRHandleReducedOCRHandleReducedOCRHandleReducedOCRHandleReducedocrhandle_reduced. The parameters
MethodMethodMethodmethodmethod, MinRemainingSVMinRemainingSVMinRemainingSVminRemainingSVmin_remaining_sv and MaxErrorMaxErrorMaxErrormaxErrormax_error have
the same meaning as in reduce_class_svmreduce_class_svmReduceClassSvmReduceClassSvmreduce_class_svm and are described
there. Please note that classification time can also be
significantly reduced with a preprocessing step in
create_ocr_class_svmcreate_ocr_class_svmCreateOcrClassSvmCreateOcrClassSvmcreate_ocr_class_svm, which possibly introduces less
errors.
Execution Information
- Multithreading type: reentrant (runs in parallel with non-exclusive operators).
- Multithreading scope: global (may be called from any thread).
- Processed without parallelization.
Parameters
OCRHandleOCRHandleOCRHandleOCRHandleocrhandle (input_control) ocr_svm → HOCRSvm, HTupleHHandleHTupleHtuple (handle) (IntPtr) (HHandle) (handle)
Original handle of SVM-based OCR-classifier.
MethodMethodMethodmethodmethod (input_control) string → HTuplestrHTupleHtuple (string) (string) (HString) (char*)
Type of postprocessing to reduce number of SVs.
Default:
'bottom_up'
"bottom_up"
"bottom_up"
"bottom_up"
"bottom_up"
List of values:
'bottom_up'"bottom_up""bottom_up""bottom_up""bottom_up"
MinRemainingSVMinRemainingSVMinRemainingSVminRemainingSVmin_remaining_sv (input_control) integer → HTupleintHTupleHtuple (integer) (int / long) (Hlong) (Hlong)
Minimum number of remaining SVs.
Default:
2
Suggested values:
2, 3, 4, 5, 7, 10, 15, 20, 30, 50
Restriction:
MinRemainingSV >= 2
MaxErrorMaxErrorMaxErrormaxErrormax_error (input_control) real → HTuplefloatHTupleHtuple (real) (double) (double) (double)
Maximum allowed error of reduction.
Default:
0.001
Suggested values:
0.0001, 0.0002, 0.0005, 0.001, 0.002, 0.005, 0.01, 0.02, 0.05
Restriction:
MaxError > 0.0
OCRHandleReducedOCRHandleReducedOCRHandleReducedOCRHandleReducedocrhandle_reduced (output_control) ocr_svm → HOCRSvm, HTupleHHandleHTupleHtuple (handle) (IntPtr) (HHandle) (handle)
SVMHandle of reduced OCR classifier.
Result
If the parameters are valid the operator
reduce_ocr_class_svmreduce_ocr_class_svmReduceOcrClassSvmReduceOcrClassSvmreduce_ocr_class_svm returns the value 2 (
H_MSG_TRUE)
. If necessary,
an exception is raised.
Possible Predecessors
trainf_ocr_class_svmtrainf_ocr_class_svmTrainfOcrClassSvmTrainfOcrClassSvmtrainf_ocr_class_svm,
get_support_vector_num_ocr_class_svmget_support_vector_num_ocr_class_svmGetSupportVectorNumOcrClassSvmGetSupportVectorNumOcrClassSvmget_support_vector_num_ocr_class_svm
Possible Successors
do_ocr_single_class_svmdo_ocr_single_class_svmDoOcrSingleClassSvmDoOcrSingleClassSvmdo_ocr_single_class_svm,
do_ocr_multi_class_svmdo_ocr_multi_class_svmDoOcrMultiClassSvmDoOcrMultiClassSvmdo_ocr_multi_class_svm,
get_support_vector_ocr_class_svmget_support_vector_ocr_class_svmGetSupportVectorOcrClassSvmGetSupportVectorOcrClassSvmget_support_vector_ocr_class_svm,
get_support_vector_num_ocr_class_svmget_support_vector_num_ocr_class_svmGetSupportVectorNumOcrClassSvmGetSupportVectorNumOcrClassSvmget_support_vector_num_ocr_class_svm
See also
create_ocr_class_svmcreate_ocr_class_svmCreateOcrClassSvmCreateOcrClassSvmcreate_ocr_class_svm
References
Bernhard Schölkopf, Alexander J.Smola: “Learning with Kernels”;
The MIT Press, London; 1999.
Module
OCR/OCV