I know there are new OcrEngineMode value in Tesseract.
But not in tess-two.
In tesseract 4.x, ocrEngineMode is :
enum OcrEngineMode {
OEM_TESSERACT_ONLY, // Run Tesseract only - fastest; deprecated
OEM_LSTM_ONLY, // Run just the LSTM line recognizer.
OEM_TESSERACT_LSTM_COMBINED, // Run the LSTM recognizer, but allow
fallback
// to Tesseract when things get difficult.
// deprecated
OEM_DEFAULT, // Specify this mode when calling init_*(),
// to indicate that any of the above modes
// should be automatically inferred from the
// variables in the language-specific
config,
// command-line configs, or if not specified
// in any of the above should be set to the
// default OEM_TESSERACT_ONLY.
OEM_COUNT // Number of OEMs
};
However, in the newest release of tess-two, the ocrEngineMode is :
@IntDef({OEM_TESSERACT_ONLY, OEM_CUBE_ONLY,
OEM_TESSERACT_CUBE_COMBINED, OEM_DEFAULT})
public @interface OcrEngineMode {}
public static final int OEM_TESSERACT_ONLY = 0;
@Deprecated
public static final int OEM_CUBE_ONLY = 1;
@Deprecated
public static final int OEM_TESSERACT_CUBE_COMBINED = 2;
public static final int OEM_DEFAULT = 3;
If there is no way to set OEM_LSTM_ONLY in tess-two,
I can only assume this is a bug in tess-two.
Quan Nguyen於 2019年12月9日星期一 UTC+8上午12時38分56秒寫道：
>
> There are new OcrEngineMode
> <https://github.com/tesseract-ocr/tesseract/blob/master/include/tesseract/publictypes.h>
>
> values.
>
>
> On Saturday, December 7, 2019 at 7:37:49 PM UTC-6, NY C wrote:
>>
>> Hi, I am using tess-two for OCR.
>>
>>
>> (Alex Chon version : https://github.com/alexcohn/tess-two
>> <https://www.google.com/url?q=https%3A%2F%2Fgithub.com%2Falexcohn%2Ftess-two&sa=D&sntz=1&usg=AFQjCNEQGm3c_HnjOOVpdOoDYCwnElOb5Q>
>> )
>>
>>
>> Code:
>>
>> TessBaseAPI baseApi = new TessBaseAPI();
>> baseApi.setDebug(true);
>> baseApi.init(pathfiles, language);
>> //baseApi.setVariable(TessBaseAPI.VAR_CHAR_WHITELIST, "0123456789");
>> baseApi.setPageSegMode(TessBaseAPI.PageSegMode.PSM_AUTO);
>> baseApi.setImage(bmp);
>> result= baseApi.getUTF8Text();
>> baseApi.end();
>>
>>
>> The code run perfectly when I use this tessdata :
>> https://github.com/tesseract-ocr/tessdata
>>
>> But when I use tessdata_fast (
>> https://github.com/tesseract-ocr/tessdata_fast), The code crashes on
>> baseApi.init.
>>
>>
>> There is no error message since the init method calls native C++. As far
>> as I can trace, the init method crashes on this line:
>>
>> boolean success = nativeInitOem(mNativeData, datapath, language,
>> ocrEngineMode);
>>
>>
>> I also tried to set the OEM like this:
>>
>> baseApi.init(pathfiles, language, TessBaseAPI.OEM_CUBE_ONLY);
>>
>>
>> All the OEM parameters have been tried :
>>
>> (OEM_TESSERACT_ONLY = 0, OEM_CUBE_ONLY = 1, OEM_TESSERACT_CUBE_COMBINED =
>> 2, OEM_DEFAULT = 3)
>>
>> Crashes as well.
>>
>>
>> How could I fix this?
>>
>>
>>
>>
--
You received this message because you are subscribed to the Google Groups
"tesseract-ocr" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/tesseract-ocr/4a0d7fba-73fe-43d9-96e7-55072b82f876%40googlegroups.com.
