Newbie question - Need to extract text out of user-generated images

[newbie]

Hi,

     I need to programtically extract text out of images(user-generated using digital camera) and use this  text to search against our equipment document repository for information. I also have trained images(images from the equipment manufacturers in pdf) to compare against if I can do image to image matching. I tried some CV(computer vision) programs to do the matching and it was suggested OCR is the way to go in my case. So is that doable ?

I read up on extracting text out of images(not to do the match) on this forum and I see there is also prep work needs to be done to make the image to be recognizable for OCR. Even before I go about prepping the image, I would like to know what  needs to be downloaded and an example of using a java api is highly appreciated(I am sure its already there, would appreciate it if you could point me to right link).

Appreciate it!