Scene Text Recognition

Detailed Description

Classes

Name	Description
class cv::text::BaseOCR	View details
class cv::text::OCRBeamSearchDecoder	OCRBeamSearchDecoder class provides an interface for OCR using Beam Search algorithm. View details
class cv::text::OCRHMMDecoder	OCRHMMDecoder class provides an interface for OCR using Hidden Markov Models. View details
class cv::text::OCRHolisticWordRecognizer	OCRHolisticWordRecognizer class provides the functionallity of segmented wordspotting. Given a predefined vocabulary , a DictNet is employed to select the most probable word given an input image. View details
class cv::text::OCRTesseract	OCRTesseract class provides an interface with the tesseract-ocr API (v3.02.02) in C++. View details

Enumerations

enum cv::text {
    cv::OCR_LEVEL_WORD,
    cv::OCR_LEVEL_TEXTLINE
}

View details

enum cv::text::classifier_type {
    cv::text::OCR_KNN_CLASSIFIER = 0,
    cv::text::OCR_CNN_CLASSIFIER = 1
}

View details

enum cv::text::decoder_mode {
    cv::text::OCR_DECODER_VITERBI = 0
}

View details

enum cv::text::ocr_engine_mode {
    cv::text::OEM_TESSERACT_ONLY,
    cv::text::OEM_CUBE_ONLY,
    cv::text::OEM_TESSERACT_CUBE_COMBINED,
    cv::text::OEM_DEFAULT
}

Tesseract.OcrEngineMode Enumeration. View details

enum cv::text::page_seg_mode {
    cv::text::PSM_OSD_ONLY,
    cv::text::PSM_AUTO_OSD,
    cv::text::PSM_AUTO_ONLY,
    cv::text::PSM_AUTO,
    cv::text::PSM_SINGLE_COLUMN,
    cv::text::PSM_SINGLE_BLOCK_VERT_TEXT,
    cv::text::PSM_SINGLE_BLOCK,
    cv::text::PSM_SINGLE_LINE,
    cv::text::PSM_SINGLE_WORD,
    cv::text::PSM_CIRCLE_WORD,
    cv::text::PSM_SINGLE_CHAR
}

Tesseract.PageSegMode Enumeration. View details

Enumeration Type Documentation

enum

#include <opencv2/text/ocr.hpp>

Enumerator:

OCR_LEVEL_WORD

OCR_LEVEL_TEXTLINE

classifier_type

enum cv::text::classifier_type

#include <opencv2/text/ocr.hpp>

Enumerator:

OCR_KNN_CLASSIFIER Python: cv.text.OCR_KNN_CLASSIFIER

OCR_CNN_CLASSIFIER Python: cv.text.OCR_CNN_CLASSIFIER

decoder_mode

enum cv::text::decoder_mode

#include <opencv2/text/ocr.hpp>

Enumerator:

OCR_DECODER_VITERBI Python: cv.text.OCR_DECODER_VITERBI

ocr_engine_mode

enum cv::text::ocr_engine_mode

#include <opencv2/text/ocr.hpp>

Tesseract.OcrEngineMode Enumeration.

Enumerator:

OEM_TESSERACT_ONLY Python: cv.text.OEM_TESSERACT_ONLY

OEM_CUBE_ONLY Python: cv.text.OEM_CUBE_ONLY

OEM_TESSERACT_CUBE_COMBINED Python: cv.text.OEM_TESSERACT_CUBE_COMBINED

OEM_DEFAULT Python: cv.text.OEM_DEFAULT

page_seg_mode

enum cv::text::page_seg_mode

#include <opencv2/text/ocr.hpp>

Tesseract.PageSegMode Enumeration.

Enumerator:

PSM_OSD_ONLY Python: cv.text.PSM_OSD_ONLY

PSM_AUTO_OSD Python: cv.text.PSM_AUTO_OSD

PSM_AUTO_ONLY Python: cv.text.PSM_AUTO_ONLY

PSM_AUTO Python: cv.text.PSM_AUTO

PSM_SINGLE_COLUMN Python: cv.text.PSM_SINGLE_COLUMN

PSM_SINGLE_BLOCK_VERT_TEXT Python: cv.text.PSM_SINGLE_BLOCK_VERT_TEXT

PSM_SINGLE_BLOCK Python: cv.text.PSM_SINGLE_BLOCK

PSM_SINGLE_LINE Python: cv.text.PSM_SINGLE_LINE

PSM_SINGLE_WORD Python: cv.text.PSM_SINGLE_WORD

PSM_CIRCLE_WORD Python: cv.text.PSM_CIRCLE_WORD

PSM_SINGLE_CHAR Python: cv.text.PSM_SINGLE_CHAR

Function Documentation

createOCRHMMTransitionsTable()

Mat cv::text::createOCRHMMTransitionsTable(
const String & vocabulary,
std::vector< cv::String > & lexicon )

#include <opencv2/text/ocr.hpp>

Python:

cv.text.createOCRHMMTransitionsTable(vocabulary, lexicon) -> retval

createOCRHMMTransitionsTable()

void cv::text::createOCRHMMTransitionsTable(
std::string & vocabulary,
std::vector< std::string > & lexicon,
OutputArray transition_probabilities_table )

#include <opencv2/text/ocr.hpp>

Python:

cv.text.createOCRHMMTransitionsTable(vocabulary, lexicon) -> retval

Utility function to create a tailored language model transitions table from a given list of words (lexicon).

The function calculate frequency statistics of character pairs from the given lexicon and fills the output transition_probabilities_table with them. The transition_probabilities_table can be used as input in the OCRHMMDecoder::create() and OCRBeamSearchDecoder::create() methods.

Note

• (C++) An alternative would be to load the default generic language transition table provided in the text module samples folder (created from ispell 42869 english words list) : opencv/opencv_contrib

Parameters

• vocabulary — The language vocabulary (chars when ASCII English text).

• lexicon — The list of words that are expected to be found in a particular image.

• transition_probabilities_table — Output table with transition probabilities between character pairs. cols == rows == vocabulary.size().

loadOCRBeamSearchClassifierCNN()

Ptr< OCRBeamSearchDecoder::ClassifierCallback > cv::text::loadOCRBeamSearchClassifierCNN(const String & filename)

#include <opencv2/text/ocr.hpp>

Python:

cv.text.loadOCRBeamSearchClassifierCNN(filename) -> retval

Allow to implicitly load the default character classifier when creating an OCRBeamSearchDecoder object.

The CNN default classifier is based in the scene text recognition method proposed by Adam Coates & Andrew NG in [Coates11a]. The character classifier consists in a Single Layer Convolutional Neural Network and a linear classifier. It is applied to the input image in a sliding window fashion, providing a set of recognitions at each window location.

Parameters

• filename — The XML or YAML file with the classifier model (e.g. OCRBeamSearch_CNN_model_data.xml.gz)

loadOCRHMMClassifier()

Ptr< OCRHMMDecoder::ClassifierCallback > cv::text::loadOCRHMMClassifier(
const String & filename,
int classifier )

#include <opencv2/text/ocr.hpp>

Python:

cv.text.loadOCRHMMClassifier(filename, classifier) -> retval

Allow to implicitly load the default character classifier when creating an OCRHMMDecoder object.

Parameters

• filename — The XML or YAML file with the classifier model (e.g. OCRBeamSearch_CNN_model_data.xml.gz)

• classifier — Can be one of classifier_type enum values.

loadOCRHMMClassifierCNN()

Ptr< OCRHMMDecoder::ClassifierCallback > cv::text::loadOCRHMMClassifierCNN(const String & filename)

#include <opencv2/text/ocr.hpp>

Python:

cv.text.loadOCRHMMClassifierCNN(filename) -> retval

Allow to implicitly load the default character classifier when creating an OCRHMMDecoder object.

The CNN default classifier is based in the scene text recognition method proposed by Adam Coates & Andrew NG in [Coates11a]. The character classifier consists in a Single Layer Convolutional Neural Network and a linear classifier. It is applied to the input image in a sliding window fashion, providing a set of recognitions at each window location.

Deprecated

use loadOCRHMMClassifier instead

Parameters

• filename — The XML or YAML file with the classifier model (e.g. OCRBeamSearch_CNN_model_data.xml.gz)

loadOCRHMMClassifierNM()

Ptr< OCRHMMDecoder::ClassifierCallback > cv::text::loadOCRHMMClassifierNM(const String & filename)

#include <opencv2/text/ocr.hpp>

Python:

cv.text.loadOCRHMMClassifierNM(filename) -> retval

Allow to implicitly load the default character classifier when creating an OCRHMMDecoder object.

The KNN default classifier is based in the scene text recognition method proposed by Lukás Neumann & Jiri Matas in [Neumann11b]. Basically, the region (contour) in the input image is normalized to a fixed size, while retaining the centroid and aspect ratio, in order to extract a feature vector based on gradient orientations along the chain-code of its perimeter. Then, the region is classified using a KNN model trained with synthetic data of rendered characters with different standard font types.

Deprecated

loadOCRHMMClassifier instead

Parameters

• filename — The XML or YAML file with the classifier model (e.g. OCRHMM_knn_model_data.xml)