setting pagesegmode for multi-column text OCR

[alexiuk]

Hi - I have an image with multiple columns I'd like to OCR. I'm using tess4j v1.1 and wrote another small convenience wrapper. 

I was expecting to be able to change pagesegmode as per the following failing junit test.  I'd sure appreciate some advice.

Thanks

@Test

public void test_setParams() {

// my convenience wrapper

  OCR_Tesseract ocr = new OCR_Tesseract();

  int current_page_seg_mode = Tesseract1.TessBaseAPIGetPageSegMode( ocr.handle );

  assertEquals(TessPageSegMode.PSM_AUTO_OSD, current_page_seg_mode);

}

// extract from ctor.

this.handle = TessAPI1.TessBaseAPICreate();

TessAPI1.TessBaseAPIInit3(this.handle, this.datapath, this.language);

this.instance = new Tesseract1();

this.instance.setPageSegMode( TessPageSegMode.PSM_AUTO_OSD );

...

[Quan Nguyen]

SetPageSegMode should be called after Init.

Take a look at the testOSD test case in http://sourceforge.net/p/tess4j/code/HEAD/tree/Tess4J_3/trunk/test/net/sourceforge/tess4j/