Android图片文字识别

导言:
OCR,tess-two ,openCV等晕人的东西先分清,OCR,tess-two是图片文字识别,而openCV是图像识别比对,对于更复杂的图片文字识别需求可以采用百度云人工智能通用文字识别开发的SDK,准确性更高

可运行的步骤:
1:添加依赖

 implementation 'com.rmtheis:tess-two:8.0.0'

2:下载字体识别库(chi_sim.traineddata 中文简体，chi_tra.traineddata 中文繁体，eng.traineddata 英文库)

3:为了apk的大小,我们需要将字体识别库文件拷贝到SD卡目录中,比如eng.traineddata的copy

  private String mDataPath = Environment.getExternalStorageDirectory().getAbsolutePath() + File.separator;
  private String mFilePath = mDataPath + File.separator + "tessdata" + File.separator + "eng.traineddata";
  private void copyFile() {
        try {
            File mFile = new File(mFilePath);
            if (mFile.exists()) {
                mFile.delete();
            }
            if (!mFile.exists()) {
                File p = new File(mFile.getParent());
                if (!p.exists()) {
                    p.mkdirs();
                }
                try {
                    mFile.createNewFile();
                } catch (IOException e) {
                    e.printStackTrace();
                }
            }

            OutputStream os = new FileOutputStream(mFilePath);
            InputStream is = this.getAssets().open("eng.traineddata");
            byte[] buffer = new byte[1024];
            int len = 0;
            while ((len = is.read(buffer)) != -1) {
                os.write(buffer, 0, len);
            }
            os.flush();
            os.close();
            is.close();
        } catch (Exception e) {
            e.printStackTrace();
        }
    }

4:tess two初始化

 TessBaseAPI baseApi;
 baseApi = new TessBaseAPI();
 baseApi.init(mDataPath, "eng");

5:处理bitmap图片并识别里面的内容

//OCR图片文字识别
baseApi.setImage(bitmap);
String result = baseApi.getUTF8Text().replace(" ", "").toLowerCase();

6:其他要求看需求,本文结束