ios身份证识别opencv+TesseractOCR

第一次写，多见谅。
本文运用opencv+TesseractOCR来实现身份证识别姓名和身份证号
查阅并引用了很多文章
opencv安装及入门
OpenCV在iOS上的应用尝试
在身份证识别上，由于相机拍摄的亮度及其他问题难点在于图片的阈值二值化的判断，用过固定阈值二值化、局部阈值二值化、平均阈值二值化都不理想。
在基于opencv的一种局部自适应快速二值化方法（积分法）文章中方法解决了光照不均匀的图像中进行二值化。

//扫描身份证图片，并进行预处理，定位号码区域图片并返回

- (NSArray*)opencvScanCard:(UIImage*)image {

//将UIImage转换成Mat

cv::MatresultImage;

UIImageToMat(image, resultImage);

//先用使用3x3内核来降噪

blur( resultImage, resultImage,cv::Size(3,3),cv::Point(-1,-1));

//UIImage *Image2 = MatToUIImage(resultImage);

////固定阈值二值化

//cv::threshold(resultImage, resultImage, 100, 255, CV_THRESH_BINARY);

////局部阈值二值化

//cv::adaptiveThreshold(resultImage, resultImage, 255, CV_ADAPTIVE_THRESH_MEAN_C, CV_THRESH_BINARY, 3, 5);

////平均阈值二值化

//IplImage imgTopDown = resultImage;

//CvScalar mean ,std_dev;//平均值、标准差

//double u_threshold,d_threshold;

//cvAvgSdv(&imgTopDown,&mean,&std_dev,NULL);

//u_threshold = mean.val[0] +1.5* std_dev.val[0];//上阀值

//d_threshold = mean.val[0] -1.5* std_dev.val[0];//下阀值

//u_threshold = mean.val[0];

//d_threshold = mean.val[0];

////u_threshold = mean + 2.5 * std_dev; //错误

////d_threshold = mean - 2.5 * std_dev;

//std::cout<<"The TopThreshold of this Image in TopDown is:"<

//std::cout<<"The DownThreshold of this Image in TopDown is:"<

//cv::threshold(resultImage,resultImage,d_threshold,u_threshold,CV_THRESH_BINARY);//上下阀值

//积分阈值二值化

resultImage =AdaptiveThereshold(resultImage, resultImage);

//腐蚀，填充（腐蚀是让黑色点变大）

cv::MaterodeElement =getStructuringElement(cv::MORPH_RECT,cv::Size(22,22));

cv::erode(resultImage, resultImage, erodeElement);

//轮廊检测

std::vector> contours;//定义一个容器来存储所有检测到的轮廊

cv::findContours(resultImage, contours,CV_RETR_TREE,CV_CHAIN_APPROX_SIMPLE,cvPoint(0,0));

//cv::drawContours(resultImage, contours, -1, cv::Scalar(255),4);

//取出身份证号码区域

std::vector rects;

cv::RectNameNumberRect =cv::Rect(0,0,0,0);

cv::RectCardNumberRect =cv::Rect(0,0,0,0);

std::vector>::const_iteratoritContours = contours.begin();

for( ; itContours != contours.end(); ++itContours) {

cv::Rectrect =cv::boundingRect(*itContours);

rects.push_back(rect);

//算法原理

if(rect.width> CardNumberRect.width&& rect.width> rect.height*6) {

CardNumberRect = rect;

}

if(rect.x>60&& rect.y<120&& rect.height*2< rect.width) {

NameNumberRect = rect;

}

}

//定位成功成功，去原图截取身份证号码区域，并转换成灰度图、进行二值化处理

//去原图截取身份证姓名区域，并转换成灰度图、进行二值化处理

cv::MatmatImage;

UIImageToMat(image, matImage);

cv::MatNameImageMat;

NameImageMat = matImage(NameNumberRect);

IplImagegrey = NameImageMat;

unsignedchar* dataImage = (unsignedchar*)grey.imageData;

intthreshold =Otsu(dataImage, grey.width, grey.height);

printf("阈值：%d\n",threshold);

NameImageMat =AdaptiveThereshold(NameImageMat, NameImageMat);

UIImage*NameImage =MatToUIImage(NameImageMat);

cv::MatCardImageMat;

CardImageMat = matImage(CardNumberRect);

CardImageMat =AdaptiveThereshold(CardImageMat, CardImageMat);

UIImage*CardImage =MatToUIImage(CardImageMat);

//身份证号码定位失败

if([publicClassobjectIsEmpty:CardImage]||[publicClassobjectIsEmpty:NameImage]) {

returnnil;

}

NSArray* arr =@[NameImage,CardImage];

returnarr;

}

局部自适应快速积分二值化方法

cv::Mat AdaptiveThereshold(cv::Mat src,cv::Mat dst)
{
    cvtColor(src,dst,CV_BGR2GRAY);
    int x1, y1, x2, y2;
    int count=0;
    long long sum=0;
    int S=src.rows>>3;  //划分区域的大小S*S
    int T=15;         /*百分比，用来最后与阈值的比较。原文：If the value of the current pixel is t percent less than this average
                       then it is set to black, otherwise it is set to white.*/
    int W=dst.cols;
    int H=dst.rows;
    long long **Argv;
    Argv=new long long*[dst.rows];
    for(int ii=0;ii(j,i);
            if(i==0)
                Argv[j][i]=sum;
            else
                Argv[j][i]=Argv[j][i-1]+sum;
        }
    }
    
    for(int i=0;i=W)
                x2=W-1;
            if(y1<0)
                y1=0;
            if(y2>=H)
                y2=H-1;
            count=(x2-x1)*(y2-y1);
            sum=Argv[y2][x2]-Argv[y1][x2]-Argv[y2][x1]+Argv[y1][x1];
            
            
            if((long long)(dst.at(j,i)*count)<(long long)sum*(100-T)/100)
                dst.at(j,i)=0;
            else
                dst.at(j,i)=255;
        }
    }
    for (int i = 0 ; i < dst.rows; ++i)
    {
        delete [] Argv[i];
    }
    delete [] Argv;
    return dst;
}

TesseractOCR识别文字

//利用TesseractOCR识别文字
- (void)tesseractRecognizeImageArr:(NSArray *)arr compleate:(CompleateBlock)compleate {
    
    dispatch_async(dispatch_get_global_queue(DISPATCH_QUEUE_PRIORITY_BACKGROUND, 0), ^{
        NSMutableArray* marr = [NSMutableArray array];
        for (int i = 0; i

参考文章：
iOS身份证号码识别
demo下载：
由于包含字体库，demo比较大，还有下载完后请自行pod install。
idCardOpencv