This is a python project for detecting and recognizing pre-defined, pre-trained objects in real timeor from a video file input. It also takes screenshots of the detected objects and save it as a bmp file.
The tools used here are : Open cv 2.4.13 Python 2.7.11 Numpy 1.9.1
Make sure you have included video.py and common.py file in your python folder.
Also rename the opencv_ffmpeg.dll file from the opencv/sources/3rdparty to opencv_ffmpeg2413.dll and copy it to the python directory inoder to read video from a file.
Now download all the haarcascade xml files provided here to your pc and plug in the location to the code.
Now i guess you are good to go!
Feel free to mess with the code... and let me know of any great modifications that you made