bill-wangbiao / ocr-tess4j-rest

OCR Tess4J - Java, Spring Boot Microservice, Optional MongoDB with maven, gradle build

ocr-tess4j-rest

ocr-tess4j-rest - Java Wrapper for Tesseract OCR with Rest API built over Tess4j (http://tess4j.sourceforge.net).

Tess4J is a JNA wrapper for Tesseract OCR API it provides character recognition support for common image formats, and multi-page images. The library has been developed and tested on Windows and Linux.

Installing dependencies

Installing Tesseract
https://code.google.com/p/tesseract-ocr/

Install ghostscript (for PDF to Text) * If you are going to do PDF to text as well*
http://www.ghostscript.com/download/gsdnld.html

*On Mac the easiest way is to use homebrew:

brew install tesseract
brew install gs

---

ocr-tess4j-rest uses:

• Spring Boot for Rest
• Spring Boot Data for connecting with mongo db (/v1 endpoint uses mongo).
• Image + Text (from OCR) is stored in mongo db for /v1 endpoint.
• Text of uploaded image is displayed on console/log if you use /v0.9 (Testing or if you want to use something else than mongo).
• Just run Tess4jV1.java and it will spin up an embedded Tomcat on localhost 8080
• Rest Assured is used for testing rest (Tess4jV1). Just remove @Ignore on the Tess4jV1SmokeTest and run the rest test.
• Logback for logging.
• Graddle for build (or)
• Maven for build.

*This version wraps tess4j as dependency and pulls it as a dependency jar and has spring boot upgraded to 1.1.9.
*Tested with JDK 1.7.72, Tesseract 3.02.02 on MAC.