I need to perform scene text recognition with OpenCV as explained in its [official documentation][1]. However, because of some inevitable technical constraints I need to do it using the Java wrapper of the library.
I have had a look at two official examples included with OpenCV: an [end to end recognition demo][2] and [one that uses the webcam][3]. The core functionality of these examples is based on the use of an `ERFilter`, which performs the actual classification given an image.
I have managed to compile OpenCV's source code (version 3.1.0) with the support of the `text module` available at the [repository of extra modules][4]. However, I have not been able to locate an `ERFilter` Java class (or any equivalent functionality) within the Java wrapper that results after my compilation process. In fact, I have not been able to find any importable option similar to `org.opencv.text`, which is my guess on where such a class might be.
Does anyone have any idea on how to do this with the Java API provided by the compiled library?
Any approach on how to *translate* the official C++ code into its corresponding Java version?
Your suggestions would be very much appreciated.
[1]: http://docs.opencv.org/3.0-beta/modules/text/doc/ocr.html
[2]: https://github.com/Itseez/opencv_contrib/blob/master/modules/text/samples/end_to_end_recognition.cpp
[3]: https://github.com/Itseez/opencv_contrib/blob/master/modules/text/samples/webcam_demo.cpp
[4]: https://github.com/Itseez/opencv_contrib
↧