[JAVA] OCR - tesseract

테서렉트의 목적

• 광학 문자 인식 OCR(Optical character recognition)
• 이미지(사진)에서 글자를 추출

테서렉트의 장점

• 무료

 

테서렉트의 단점

• 한글 인식률이 생각보다 저조하다

구현

<!-- https://mvnrepository.com/artifact/net.sourceforge.tess4j/tess4j -->
<dependency>
	<groupId>net.sourceforge.tess4j</groupId>
	<artifactId>tess4j</artifactId>
	<version>4.5.4</version>
</dependency>

package spider.binaries.app.util;

import java.io.File;

import net.sourceforge.tess4j.Tesseract;
import net.sourceforge.tess4j.TesseractException;

public class TesseractTest {

	public static void main(String[] args) {
		String result = ocrImage("PATH");
		System.out.println(result);
	}
	
	
	private static Tesseract getTesseract() {
		
		Tesseract instance = new Tesseract();
		instance.setDatapath("/Users/devs/Downloads/tessdata");
		instance.setLanguage("kor");//"kor+eng"
		return instance;
		
	}

	private static String ocrImage(String fileName) {
		
		Tesseract tesseract = getTesseract();
		
		String result = null;
		
		File file = new File("/Users/devs/Downloads/tessdata/download/"+fileName);
		
		if(file.exists() && file.canRead()) {
			try {
				result = tesseract.doOCR(file);
			} catch (TesseractException e) {
				result = e.getMessage();
			}	
		} else {
			result = "not exist";
		}
		return result;

	}
		
}

 

설치시 주의 사항

 

 

출처 : https://github.com/tesseract-ocr/tesseract

          https://tesseract-ocr.github.io/

          https://en.wikipedia.org/wiki/Tesseract_(software)

tesseract-ocr/tesseract

 

Tesseract documentation

 

Tesseract (software) - Wikipedia