Large volume handwriting recognition of numbers
This project received 5 bids from talented freelancers with an average bid price of $4021 USD.Get free quotes for a project like this
I have several million documents that have ten digit numbers written at the bottom of the first page. I need a program that can do handwriting recognition to read these numbers.
Only some of the documents have this number, so the system needs to be able to recognize when a document doesn't have a number.
I would prefer a system that can return more than one alternative for each digit and a probability for each alternative. (ie, the third digit has a 56% probability of being a 3, and 23% probability of being an 8, a 9% probability of being a 2, and a 2% probability of being something else). I have other sources of information that I can use to narrow the field of possibilities.
I would like a standalone program that can process a batch of documents, for example given a directory, output a list of the filenames and the recognition results. I will undertake to spread the program over multiple machines and create the batches.
Code can be written in any language and can use external libraries as long as the are open source. Code can run on Windows, Linux, or FreeBSD.
The code should be well written and optimized for speed.
An example of a document can be found at this URL. Look at the 10 digit number at the bottom.
<[url removed, login to view]>
1) Complete and fully-functional working program(s) in executable form as well as complete source code of all work done.
2) Deliverables must be in ready-to-run condition, as follows (depending on the nature of the deliverables):
a) For web sites or other server-side deliverables intended to only ever exist in one place in the Buyer's environment--Deliverables must be installed by the Seller in ready-to-run condition in the Buyer's environment.
b) For all others including desktop software or software the buyer intends to distribute: A software installation package that will install the software in ready-to-run condition on the platform(s) specified in this bid request.
3) All deliverables will be considered "work made for hire" under U.S. Copyright law. Buyer will receive exclusive and complete copyrights to all work purchased. (No GPL, GNU, 3rd party components, etc. unless all copyright ramifications are explained AND AGREED TO by the buyer on the site per the coder's Seller Legal Agreement).
* * *This broadcast message was sent to all bidders on Wednesday Sep 17, 2008 4:48:01 AM:
I've attached a .zip file with some samples of the handwritten numbers.
First, thank you for your interest in this project. I hope you will see that I have done a large number of projects on RAC, that I have maintained excellent ratings, *and* that I have a very low non-action ratio.
Furthermore, I will probably choose 2-3 coders to work on this project in parallel. I'm sure you recognize that this is the type of project where different programs can be successful at different documents.
Finally, it seems that fewer than 10% of my documents have the hand-written number at the bottom. That means that for a high-volume approach it would be good if the programs were to detect quickly when there isn't a number written. That way they can spend more time on the documents that do have a number.
Windows, Linux, or FreeBSD
Any programming language
Looking to make some money?
- Set your budget and the timeframe
- Outline your proposal
- Get paid for your work
Hire Freelancers who also bid on this project
Looking for work?
Work on projects like this and make money from home!Sign Up Now
- The New York Times
- Wall Street Journal
- Times Online