tesseract android环境搭建

环境
1.mac os
2.android studio 3.1.3
3.tess-two 8.0.0
下载地址: https://github.com/rmtheis/tess-two

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

注：tesseract没有提供详细的android接口，这里使用tess-two

1.build.gradle引入tess-two

dependencies {
    implementation 'com.rmtheis:tess-two:8.0.0'
}

2.导入识别库 *.traineddata

3.加载识别库

    package cn.xylink.ocr.image.cn.xylink.ocr.image.config;

import android.content.Context;
import android.os.Environment;
import android.util.Log;

import com.googlecode.tesseract.android.TessBaseAPI;

import java.io.File;
import java.io.FileOutputStream;
import java.io.InputStream;

import cn.xylink.ocr.image.R;

/**
 * @author hecj
 */
public class TessBaseAPIFactory {

    static TessBaseAPI tessApi;

    static String datapath= Environment.getExternalStorageDirectory().getAbsolutePath()+ "/tesseract";
    static final String DEFAULT_LANGUAGE = "xylinkfont";

    public static TessBaseAPI getTessBaseAPIInstance(Context context){
        try {
            tessApi = new TessBaseAPI();
            File dir = new File(datapath);
            tessApi.setDebug(true);
            if (!dir.exists()) {
                dir.mkdirs();
            }
            File tessdatadir = new File(datapath+"/tessdata");
            if(!tessdatadir.exists()){
                tessdatadir.mkdirs();
            }
            InputStream input = context.getResources().openRawResource(R.raw.xylinkfont);
            File file = new File(dir+"/tessdata", "xylinkfont.traineddata");
            FileOutputStream output = new FileOutputStream(file);
            byte[] buff = new byte[1024];
            int len = 0;
            while ((len = input.read(buff)) != -1) {
                output.write(buff, 0, len);
            }
            input.close();
            output.close();
            boolean success = tessApi.init(datapath, DEFAULT_LANGUAGE);
            if (success) {
                Log.i("tess", "load Tesseract OCR Engine successfully...");
            } else {
                Log.i("tess", "WARNING:could not initialize Tesseract data...");
            }
        } catch (Exception ex){
            ex.printStackTrace();
        }
        return tessApi;
    }
}

使用：

 /**
     * 识别
     */
    private void ocr (){
        String resultText = "";
        try {
            Bitmap bitmap =((BitmapDrawable) ((ImageView) imageView).getDrawable()).getBitmap();
            TessBaseAPI tessApi = TessBaseAPIFactory.getTessBaseAPIInstance(context);
            tessApi.setImage(bitmap);
            // 识别结果
            resultText = tessApi.getUTF8Text();
            tessApi.end();
        }catch (Exception ex){
            Log.e("tess",ex.getMessage());
            ex.printStackTrace();
        }finally {
            text.setText("识别结果:"+resultText);
        }
    }