OpenCV二进制自适应阈值OCR

时间:2014-04-24 05:14:00

标签: c++ opencv binary-data adaptive-threshold

我需要将一些图像转换为二进制文件以进行OCR。

以下是我正在使用的功能:

Mat binarize(Mat & Img, Mat& res, float blocksize, bool inverse)
{
    Img.convertTo(Img,CV_32FC1,1.0/255.0);
    CalcBlockMeanVariance(Img,res, blocksize, inverse);
    res=1.0-res;
    res=Img+res;
    if (inverse) {
        cv::threshold(res,res,0.85,1,cv::THRESH_BINARY_INV);
    } else {
        cv::threshold(res,res,0.85,1,cv::THRESH_BINARY);
    }
    cv::resize(res,res,cv::Size(res.cols/2,res.rows/2));

    return res;
}

CalcBlockMeanVariance

void CalcBlockMeanVariance(Mat& Img,Mat& Res,float blockSide, bool inverse) //21 blockSide - the parameter (set greater for larger font on image)
{
    Mat I;
    Img.convertTo(I,CV_32FC1);
    Res=Mat::zeros(Img.rows/blockSide,Img.cols/blockSide,CV_32FC1);
    Mat inpaintmask;
    Mat patch;
    Mat smallImg;
    Scalar m,s;

    for(int i=0;i<Img.rows-blockSide;i+=blockSide)
    {
        for (int j=0;j<Img.cols-blockSide;j+=blockSide)
        {
            patch=I(Range(i,i+blockSide+1),Range(j,j+blockSide+1));
            cv::meanStdDev(patch,m,s);
            if(s[0]>0.01) // Thresholding parameter (set smaller for lower contrast image)
            {
                Res.at<float>(i/blockSide,j/blockSide)=m[0];
            }else
            {
                Res.at<float>(i/blockSide,j/blockSide)=0;
            }
        }
    }

    cv::resize(I,smallImg,Res.size());

    if (inverse) {
        cv::threshold(Res,inpaintmask,0.02,1.0,cv::THRESH_BINARY_INV);
    } else {
        cv::threshold(Res,inpaintmask,0.02,1.0,cv::THRESH_BINARY);
    }


    Mat inpainted;
    smallImg.convertTo(smallImg,CV_8UC1,255);

    inpaintmask.convertTo(inpaintmask,CV_8UC1);
    inpaint(smallImg, inpaintmask, inpainted, 5, INPAINT_TELEA);

    cv::resize(inpainted,Res,Img.size());
    Res.convertTo(Res,CV_32FC1,1.0/255.0);

}

当我将1传递给CalcBlockMeanVariance blockSide时,我得到了这个结果,我试图提升blockSide,但这只会导致更糟糕的结果。

在:

enter image description here

后:

enter image description here

有人可以建议使用不同的方法将此图像转换为二进制文件作为OCR的准备工作吗?

感谢。

1 个答案:

答案 0 :(得分:7)

我认为您可以使用Otsu方法进行阈值处理。您可以将它应用于整个图像或图像块。我做了以下步骤:

  • 使用Otsu方法对所需输入进行阈值处理。
  • Closing结果。

Python代码

image = cv2.imread('image4.png', cv2.IMREAD_GRAYSCALE) # reading image
if image is None:
    print 'Can not find the image!'
    exit(-1)
# thresholding image using ostu method
ret, thresh = cv2.threshold(image, 0, 255, cv2.THRESH_BINARY_INV | cv2.THRESH_OTSU) 
# applying closing operation using ellipse kernel
N = 3
kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (N, N))
thresh = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, kernel)
# showing the result
cv2.imshow('thresh', thresh)
cv2.waitKey(0)
cv2.destroyAllWindows()

<强>解释

在第一部分中,我使用imread读取输入图像并检查图像是否正确打开!

image = cv2.imread('image4.png', cv2.IMREAD_GRAYSCALE) # reading image
if image is None:
    print 'Can not find the image!'
    exit(-1)

现在使用otsu方法以thresh方式作为参数,使用THRESH_BINARY_INV | THRESH_OTSU方法对图像进行阈值处理。 otsu方法基于优化问题,找到阈值的最佳值。因此,我通过给出0的下限和255的上限来提供阈值的可能值范围。

ret, thresh = cv2.threshold(image, 0, 255, cv2.THRESH_BINARY_INV | cv2.THRESH_OTSU)

使用Ellipse内核完成关闭图像去除黑洞的操作。

kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (N, N))
thresh = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, kernel)

<强>结果

Figure 1