I am trying to make an OCR library using Haar classifier of opencv.But its not working well.So can you tell me what are the steps for making an OCR? and will it be possible by using Haar classifier?
There are lots of books, tutorials etc available on googling.
But there are two books which would be useful for you.
1) Handbook of Character Recognition and Document Image Analysis by H. Bunke and P. S. P. Wang
2) Optical Character Recognition: An Illustrated Guide to the Frontier (The Springer International Series in Engineering and Computer Science) by Stephen V. Rice, George Nagy and Thomas A. Nartker
They have got some good deep theory in it.
And you can see a simple digit OCR using kNN in C++ here: Basic OCR in OpenCV
And you can see a simple Python implementation of OCR using kNN in this SOF: Simple Digit Recognition OCR in OpenCV-Python
In the OpenCV svn, you can see a new Python sample for hand-written recognition using ANN : ...svn\opencv\samples\python2\digits.py . The code is in Python, but all OpenCV functions would be available in C++ too.
Hope this would be sufficient for a start up.
Related
Blockquote
i am working on this project asssigned by university as final project. But the issue is i am not getting any help from the internet so i thought may be asking here can solve issue. i had read many articles but they had no code or guidance and i am confused what to do. Basically it is an image processing work with machine learning. Data set can be found easily but issue is python python learning algorithm and code
Blockquote
I presume if it's your final project you have to create the program yourself rather than ripping it straight from the internet. If you want a good starting point which you can customise Tensor Flow from Google is very good. You'll want to understand how it works (i.e. how machine learning works) but as a first step there's a good example of image processing on the website in the form of number recognition (which is also the "Hello World" of machine learning).
https://www.tensorflow.org/get_started/mnist/beginners
This also provides a good intro to machine learning with neural nets: https://www.youtube.com/watch?v=uXt8qF2Zzfo
One note on Tensor Flow, you'll probably have to use Python 3.5+ as in my experience it can be difficult getting it on 2.7.
First of all I need to know what type of data are you using because depending on your data, if it is a MRI or PET scan or CT, there could be different suggestion for using machine learning in python for detection.
However, I suppose your main dataset consist of MR images, I am attaching an article which I found it a great overview of different methods>
This project compares four different machine learning algorithms: Decision Tree, Majority, Nearest Neighbors, and Best Z-Score (an algorithm of my own design that is a slight variant of the Na¨ıve Bayes algorithm)
https://users.soe.ucsc.edu/~karplus/abe/Science_Fair_2012_report.pdf
Here, breast cancer and colorectal cancer have been considered and the algorithms that performed best (Best Z-Score and Nearest Neighbors) used all features in classifying a sample. Decision Tree used only 13 features for classifying a sample and gave mediocre results. Majority did not look at any features and did worst. All algorithms except Decision Tree were fast to train and test. Decision Tree was slow, because it had to look at each feature in turn, calculating the information gain of every possible choice of cutpoint.
My Solution:-
Lung Image Database Consortium provides open access dataset for Lung Cancer Images.
Download it then apply any machine learning algorithm to classify images having tumor cells or not.
I attached a link for reference paper. They applied neural network to classify the images.
For coding part, use python "OpenCV" for image pre-processing and segmentation.
When it comes for classification part, use any machine learning libraries (tensorflow, keras, torch, scikit-learn... much more) as you are compatible to work with and perform classification using any better outperforming algorithms as you wish.
That's it..
Link for Reference Journal
Sorry this must be probably a dumb question. but i am fairly new to machine learning and Tessaract OCR. I have heard that Tessaract OCR can be trained.
What i need to know is does Tessaract OCR uses neural networks as their default training mechanism or do we have to program it explicitly to use neural networks ?.
Sorry if i'm thinking in a wrong way about this "training" concept. but what i need to know exactly is is Tessaract already using NN or if not how i can approach using NN with tessaract OCR to improve recognition accuracy ?.
If one can please suggest me some good resources/way to refer/try and to get started it would be a great help too.
what i currently know about basic machine learning supervised training concept and to perform basic image OCR operation in Tessaract OCR.
It appears that Tessaract uses an Adaptive Classifier by default. Check this out for a good read:
https://github.com/tesseract-ocr/docs/blob/master/tesseracticdar2007.pdf
There appears to be an option called "Cube mode" where it will switch to using NNs for the learning system instead of the adaptive classifier (https://code.google.com/p/tesseract-ocr-extradocs/wiki/Cube). More info about adaptive classifiers:
http://www.cs.indiana.edu/~rawlins/website/adaptivity/information-helper.html
Also, related very closely is a Learning Classifier System:
http://en.wikipedia.org/wiki/Learning_classifier_system
Also, your terminology of "training" is very close. Training is how you teach the pattern recognition system or learning system what responses it should give to certain input sets. Then, it uses similarities when it encounters unknown data to classify the new data. Machine learning is one of the coolest fields in existence in my opinion (probably biased opinion but whatever!) keep up the learning! You are the meta learner: learning how to teach a machine to learn! Cool stuff!
Yes, starting from tesseract 4.0, it provides a new lstm-based ocr engine: https://tesseract-ocr.github.io/tessdoc/NeuralNetsInTesseract4.00
I am trying to implement an efficient algorithm for nearest-neighbour search problem.
I have read tutorials about some data structures, which support operations for this kind of problems (for example, R-tree, cover tree, etc.), but all of them are difficult to implement.
Also I cannot find sample source code for these data structures. I know C++ and I am trying to solve this problem in this language.
Ideally, I need links that describe how to implement these data structures using source code.
There are several good choices of fast nearest neighbor search libraries.
ANN, which is based on the work of Mount and Arya. This work is documented in a paper by S. Arya and D. M. Mount. "Approximate nearest neighbor queries in fixed dimensions". In Proc. 4th ACM-SIAM Sympos. Discrete Algorithms, pages 271–280, 1993.
FLANN, which is based on the work of Marius Muja & Co. There is a paper by Marius Muja and David G. Lowe, "Fast Approximate Nearest Neighbors with Automatic Algorithm Configuration", in International Conference on Computer Vision Theory and Applications (VISAPP'09), 2009. The code for FLANN is available on github
FLANN seems to be quicker in some cases, and is a more modern version of the code with solid bindings for a number of other languages, that can incorporate changes rapidly. ANN is probably a good choice if you want a solid well-tested standard library.
Edit in Response to Comment
Both of these libraries have extensive documentation and examples.
Sample code for ANN is available in the Manual, In section 2.1.4
Sample code for FLANN is available in the FLANN repository examples directory, for example /examples/flann_examples.c
You could try a linesweep algorithm to find the closest pair of points: http://community.topcoder.com/tc?module=Static&d1=tutorials&d2=lineSweep.
Does anyone know if there is an open source implementation of oriented Basic Image Features (oBIF)? There's talk about it at http://www.cs.ucl.ac.uk/staff/m.lillholm/or.html and http://blog.kaggle.com/2011/05/04/andrew-newell-and-lewis-griffin-on-winning-the-icdar-2011-competition/. I looked through OpenCV some, but I didn't see anything that looked like this. But maybe I just don't understand all the terminology?
If not, I'd like to try to understand how hard it would be to implement. But I'm new to machine vision, and I'm having trouble understanding the image-specific jargon at the links above ('saddle-like', '2nd order gaussian filter bank', etc - I do know what a Gaussian is -just not a 2nd order filter bank of Gaussian's). What's a good reference for learning these machine vision concepts?
Full disclosure: I would like to know how to use them to compete in this years Kaggle competition (http://www.kaggle.com/c/awic2012) for Arabic Handwriting Recognition. I have more of a Machine Learning background, but I'm interested in using the features that won last years competition.
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 7 years ago.
Improve this question
I am trying to do some Face Recognition (not detection) stuff using OpenCV. I found this article with some code:
http://www.cognotics.com/opencv/servo_2007_series/index.html
However, this code is written using the older C-style OpenCV API. Does someone have a C++ API version of this using a more recent version like OpenCV 2.3.1?
Update: OpenCV 2.4.2 now comes with the very new cv::FaceRecognizer. Please see the very detailed documentation at:
http://docs.opencv.org/trunk/modules/contrib/doc/facerec/index.html
I have released libfacerec, a modern face recognition library for the OpenCV C++ API (BSD license). libfacerec has no additional dependencies and implements the Eigenfaces method, Fisherfaces method and Local Binary Patterns Histograms. Parts of the library are going to be included in OpenCV 2.4.
The latest revision of the libfacerec is available at:
https://github.com/bytefish/libfacerec
The library was written for OpenCV 2.3.1 with the upcoming OpenCV 2.4 in mind, so I don't support OpenCV versions earlier than 2.3.1. This project comes as a CMake project with a well-documented API, there's also a tutorial on gender classification. You can see a HTML version of the documentation at:
http://www.bytefish.de/dev/libfacerec/
If you want to understand how those algorithms work, you might want to read my Guide To Face Recognition (includes Python and GNU Octave/MATLAB examples):
http://www.bytefish.de/blog/face_recognition_with_opencv2
The relevant publications are:
Turk, M., and Pentland, A. Eigenfaces for recognition.. Journal of Cognitive Neuroscience 3 (1991), 71–86.
Belhumeur, P. N., Hespanha, J., and Kriegman, D. Eigenfaces vs. Fisherfaces: Recognition using class specific linear projection.. IEEE Transactions on Pattern Analysis and Machine Intelligence 19, 7 (1997), 711–720.
Ahonen, T., Hadid, A., and Pietikainen, M. Face Recognition with Local Binary Patterns.. Computer Vision - ECCV 2004 (2004), 469–481.
I'm doing a face recognition project for my engineer's degree, using c++ api. I think that everything regarding face recognition in c++ is fairly straightforward, even simpler than in C (less pointers). To use PCA you have a class named PCA described here. Just use the proper methods and read documentation with understanding. To build the matrix with input data I've created a matrix of proper size, then pasted pictures as rows (use method reshape) into it (there is a method in cv::Mat that lets you to get easily a row of a matrix). You just need to keep sure that base data and tested data have the same parameters (channels, size,etc.)
EDIT:
using namespace cv; //somewhere near top
inserting data to data matrix:
62 Mat reshaped=img.reshape(1,1);
63 Mat dataRow=_data.row(y++);
64 resize(reshaped,dataRow,dataRow.size(),0,0,CV_INTER_LINEAR);
computing pca:
251 _pca(_data,Mat(),CV_PCA_DATA_AS_ROW); //compute pca
252 _pca.project(_data,_vectors); // project original data to new coordinates
As opencv's documentation isn't the best out there, it doesn't hurt to spend some time reading it. Most of the c api functions have their equivalents in c++ api, You only need to do some "write into search window and hit enter" searching. And, there are also tutorials in c++ to get a grip of the c++ api.