tesseract源码Page Layout解读1（ 二值化，otsu）

代码插入：

在代码中加入下面一行，在tesseract/win64/bin/Realease/可以得到二值化后的图像（tessinput.tif）

api.SetVariable("tessedit_write_images", "true"); 

tessinput.tif（otsu得到的）

代码理解：

TessBaseAPI::ProcessPage[api/tesseractmain.cpp]       ->  
TessBaseAPI::Recognize [api/baseapi.cpp] -> 
/*
//Recognize the tesseract global image and return the result as Tesseractinternal structures.
int TessBaseAPI::Recognize(ETEXT_DESC* monitor) {
    ...
    if (FindLines() != 0)
    return -1;
    ...
    }
*/
TessBaseAPI::FindLines[api/baseapi.cpp] -> 
/** Find lines from the image making the BLOCK_LIST. 
int TessBaseAPI::FindLines() {
    ...
    if (tesseract_->pix_binary() == NULL && //
      !Threshold(tesseract_->mutable_pix_binary())) {//进入二值化
    return -1;
//if (tesseract_->tessedit_write_images) 
//pixWrite("pix_binary.tif", tesseract_->pix_binary(), IFF_TIFF_G4);//可以查看二值化的结果
      
	...
    }
//pix_binary_:初始值为输入的图像，保存布局分析和识别的图像
//Image used for input to layout analysis and tesseract recognition.May be modified by the ShiroRekhaSplitter to eliminate the top-line.
//mutable_pix_binary:&pix_binary_(Destroy any existing pix and return a pointer to the pointer.) 
*/
TessBaseAPI::Threshold[api/baseapi.cpp] ->
/**二值化
 * Run the thresholder to make the thresholded image, returned in pix,
 * which must not be NULL. *pix must be initializ