subokita/robust-text-detection

Robust Text Detection implementation based on http://www.stanford.edu/~hchen2/papers/ICIP2011_RobustTextDetection.pdf

This is my upcoming project: https://plusplusone.herokuapp.com

Please register your email address if you're interested in it.

---

Robust-Text-Detection

Robust Text Detection implementation based on Chen, Huizhong, et al. "Robust Text Detection in Natural Images with Edge-Enhanced Maximally Stable Extremal Regions." Image Processing (ICIP), 2011 18th IEEE International Conference on. IEEE, 2011.

which you can find the original literature here: http://www.stanford.edu/~hchen2/papers/ICIP2011_RobustTextDetection.pdf It is also partly based on Automatically Detect and Recognize Text in Natural Images example which is available here: http://www.mathworks.de/de/help/vision/examples/automatically-detect-and-recognize-text-in-natural-images.html#zmw57dd0e829

This implementation is partly motivated by the fact that helperGrowEdges and helperStrokeWidth functions on Matlab which aren't openly available to the public or non latest Matlab owner, thus those 2 functions are implemented from the scratch based on the literature and some of my own assumptions (e.g. how many pixels to prune, etc)

Feel free to correct my code, if you spotted the mistakes

robust-text-detection's Issues

Why does it have trouble with small fonts?

Hi there, this is a really good text detector. However, why does it have trouble with small font size?
For example, in this image some characters are missed.

Are there any tuning parameters I can change to help it along?

Output debug images

Saburo,
Nice job :)

Could you output debug images(cv::imwrie or cv::imshow) on each major step of the algorithm in RobustTextDetection::apply like we see in Matlab version?

Thank you in advance.

Usage license for the code?

Thanks for your efforts on this library - we'd love to try it in a project, but there's no license associated with it. Are you willing to explicitly release it under a common open source license by adding that into the root directory / readme?

Support Multiple Bounding Boxes Enclosing Text Regions

Saburo,
Could you add support of Multiple Bounding Boxes Enclosing Text Regions like in Matlab version (step6, part2)?
As far as I undetstand this will require changes of the lines 115-131 of RobustTextDetection.cpp.

Thank you in advance.

Error when building

When debug to this function, it shows the error "Access violation reading location 0xCDCDCDD1"
Mat RobustTextDetection::createMSERMask(Mat& grey) {
/* Find MSER components */
vector<vector> contours;
MSER mser(8, param.minMSERArea, param.maxMSERArea, 0.25, 0.1, 100, 1.01, 0.03, 5);
mser(grey, contours);

/* Create a binary mask out of the MSER */
Mat mser_mask(grey.size(), CV_8UC1, Scalar(0));

for (size_t i = 0; i < contours.size(); i++) {
    for (Point& point : contours[i])
        mser_mask.at<uchar>(point) = 255;
}

return mser_mask;

}

errors when compiling. error: ‘istream_iterator’ was not declared in this scope

I use opencv2.4 on my ubuntu 14.04. I just installed tesseract using "sudo apt-get install libtesseract-dev". I compiled main.cpp and I am getting the error as shown in the screenshot
kindly help!

Error for rectangular images

For rectangular images I always see this error:

Error in file "c:/2.PNG" processing: ........\opencv\modules\core\src\matrix.cpp:323: error: (-215) 0 <= roi.x && 0 <= roi.width && roi.x + roi.width <= m.cols && 0 <= roi.y && 0 <= roi.height && roi.y + roi.height <= m.rows in function cv::Mat::Mat

Example of the image:

cant get MSER class

cant you explain me where i must pust MSER class include on my code?? i read your code but i cant get it where is MSER class at all thanks