Learn something new every day
More Info... by email
Open source optical character recognition (OCR) software is a computer program that takes an image file with text and converts it into a text file, allowing users to scan written or typed documents into text documents, not just image files. To do this, the open source OCR software looks through its database of text styles and interprets the document into a text file. Choosing the best OCR program requires looking at how many text styles the program understands and its overall accuracy in guessing letters. Having a large number of interpretable image files also is useful, as is having a learning mechanism so the software can perform self-correction.
When open source OCR software sees an image file with text, such as a scanned document, the program looks simultaneously at the image file and at its text style databases. When the program sees a character it recognizes, or a similar character, it interprets that as a letter. To make the best guesses, and to increase the amount of font styles the OCR program understands, having a program with an extensive database of styles is the best. If it does not have an extensive database, the ability to add custom fonts to the program can make up for this.
While it would be good if all open source OCR software could write the correct text with 100 percent accuracy, this is not always the case. In basic terms, all OCR programs guess at characters and try to form intelligible sequences of letters and words that it thinks best interpret the document. Getting the highest accuracy OCR system will be best for the user, because less time will be spent correcting inaccurate words or phrases.
To interpret an image file with text in it, open source OCR software must support that image file. If there is no support for the image file, then it will be unable to look at it, which may dampen the program's efficiency, especially if the user has a large number of unsupported image types. Using an OCR program with the largest amount of supported file types will ensure that users will be able to have a large number of documents interpreted.
One of the major concepts behind open source OCR software is artificial intelligence (AI). This AI system is able to help the OCR program perform guesses and, after reading a new style for a time, the OCR program’s accuracy will begin to increase. Having powerful AI will introduce a self-correcting mechanism that will help accuracy without the user having to do anything.
@Soulfox -- That may be true, but that does not mean that particular situation will last forever. There are some great open source office suite, graphics programs, video editing program, pro level music mixers and recorders and other packages available.
There is no reason to assume the same will not happen to OCR programs. Heck, the Linux community thrives off of developing great open source software and making it available for as many operating systems as possible.
Unfortunately, open source OCR software is hard to find. Even if you find it, the chances that it will be as accurate as a commercial OCR reader is slim to none.
That is because the technology driving OCR programs is fairly expensive to develop. That being the case, virtually all of the top notch programs will cost some money because the developer needs to recoup a lot of costs.
One of our editors will review your suggestion and make changes if warranted. Note that depending on the number of suggestions we receive, this can take anywhere from a few hours to a few days. Thank you for helping to improve wiseGEEK!