Add OCR Service

[dannylamb]

Depends on #387

Add a simple OCR web service that executes Tesseract on any TIFF that gets POSTed to it. It should respect a header such as X-Islandora-Ocr-Args so you can pass in command line arguments.

Service will consume image/tiff.
Service will produce text/plain, text/html/ and application/pdf (because that's what Tesseract produces).

[DiegoPino]

Do you have some planning/sketching on this? Want to use something from aaron?
I do have a camel example on thumbnail creation for jpeg. maybe we could expand on that. Thanks!

[dannylamb]

I've got a skeleton that's wired up to the gradle build machinery that I'll clean up and provide after I clear out stuff for #387

[dannylamb]

We'll talk about it's mini API first, before going anywhere with it.

[whikloj]

Here is what I am using to process Tiffs to HOCR right now.
https://github.com/uml-digitalinitiatives/derivative-route/blob/master/src/main/java/ca/umanitoba/dam/DerivativeRoute.java#L245-L265

[dannylamb]

Yeah pretty much. For a web service we should stream to stdin and receive from stdout, tacking the args from a header at the end.

[ruebot]

This might be worth investigating:
http://tesseract.projectnaptha.com/
https://github.com/naptha/tesseract.js

[ruebot]

Resolved with https://github.com/Islandora-CLAW/Hypercube/commit/d3b5592ae1c4916ea39e2d6c93a3455de58830b0