Pi with openCV and Tesseract
Posted: Thu Feb 13, 2014 8:17 am
I want to dabble with video object analysis. Please help me get set up and started. I am new to these languages but figured at $35 a Raspberry Pi could do it - even if the image has to be split for 2 parallel Pi's or 2 serial Pi's to perform different kinds of analysis to get real time results. So far I have only one Pi and a 5 megapixel security camera on order.
Here's what I did so far:
Browsing pages to learn how to use webcams and security cams with Pi, I made Terminal entries to provide JPEG support and entered "sudo apt-get install python-opencv" which worked. Then in IDLE I started entering the code at http://www.jperla.com/blog/post/capturi ... m-on-linux which went fine until I entered "import opencv" which returned this in red:
Traceback (most recent call last);
File "<pyshell#0>", line 1, in <module>
import opencv
ImportError: No module named opencv
I've also read on a 3 year old comment that opencv may not be able to tell the camera to switch states so it will deliver 5 MP and not 3 MP or less. If so, what should I use?
The next question is on Tesseract OCR [url]http:/code.google.com/p/tesseract-ocr/[/url]
to identify text in the field of view. How do I hook it up? Do I need to compile it under Pi's C++ and if so, how? Do I need Boost http://www.boost.org/doc/libs/1_55_0/libs/python/doc/ or some other system to call C++ from Python?
Thank you very much.
Here's what I did so far:
Browsing pages to learn how to use webcams and security cams with Pi, I made Terminal entries to provide JPEG support and entered "sudo apt-get install python-opencv" which worked. Then in IDLE I started entering the code at http://www.jperla.com/blog/post/capturi ... m-on-linux which went fine until I entered "import opencv" which returned this in red:
Traceback (most recent call last);
File "<pyshell#0>", line 1, in <module>
import opencv
ImportError: No module named opencv
I've also read on a 3 year old comment that opencv may not be able to tell the camera to switch states so it will deliver 5 MP and not 3 MP or less. If so, what should I use?
The next question is on Tesseract OCR [url]http:/code.google.com/p/tesseract-ocr/[/url]
to identify text in the field of view. How do I hook it up? Do I need to compile it under Pi's C++ and if so, how? Do I need Boost http://www.boost.org/doc/libs/1_55_0/libs/python/doc/ or some other system to call C++ from Python?
Thank you very much.