9-1-1車牌切割

程式說明:

此車牌辨識的專案是參考底下網址matering opencv第五章，並且沿用其中的一些資料庫，
目前這邊是將其客製化成台灣的車牌，並且為了一一說明程式，將其拆成5個小專說明。
參考網址: https://github.com/MasteringOpenCV

運行結果:

這邊我們先對圖像做邊緣偵測，並且將其二值化，二值化完後用closing來連結圖塊，
利用Contour找出物體輪廓，並判斷pixel數目不足於多少時或高於多少時剃除，保留下來的為mask，
以此mask再將rgb的圖用洪水法將其圖塊塗滿，並裁剪下來，以便下專案可以做文字切割。

mask

範例程式:

#include <iostream>
#include <opencv2/opencv.hpp>
#include <opencv2/highgui/highgui.hpp>
#include <vector>
#include <algorithm>

bool verifyCarPlateSizes(cv::RotatedRect mAr)
{
float error = 0.4; //車牌面積比例的誤差
//一般台灣車牌的比例，初估為2.6
float aspect = 2.6;
//多大的面積pxiel忽略不偵測
int min = 15 * aspect * 15;
int max = 125 * aspect * 125;

float rmin = aspect - aspect*error;
float rmax = aspect + aspect*error;

int area = mAr.size.height * mAr.size.width; //計算面積

float r = (float)mAr.size.width / (float)mAr.size.height;
if (r<1)
   r = (float)mAr.size.height / (float)mAr.size.width;

if ((area < min || area > max) || (r < rmin || r > rmax))
{
   return false;
}
else
{
   return true;
}

}

cv::Mat histeq(cv::Mat in)
{
cv::Mat out(in.size(), in.type());
if (in.channels() == 3){
   cv::Mat hsv;
   std::vector hsvSplit;
   cv::cvtColor(in, hsv, CV_BGR2HSV);
   cv::split(hsv, hsvSplit);
   cv::equalizeHist(hsvSplit[2], hsvSplit[2]);
   cv::merge(hsvSplit, hsv);
   cv::cvtColor(hsv, out, CV_HSV2BGR);
}
else if (in.channels() == 1)
{
   cv::equalizeHist(in, out);
}

return out;
}

void carPlateSegment(cv::Mat& carImg, cv::Mat& carImgColor, int& num)
{

//先除雜訊
cv::blur(carImg, carImg, cv::Size(5, 5));

//先做sobel找出邊緣
cv::Mat sobleImg;

cv::Sobel(carImg, sobleImg, CV_8U, 1, 0, 3, 1, 0, cv::BORDER_DEFAULT);

//將偵測出邊緣的圖片二值化
cv::Mat binartEdgeImg;
cv::threshold(sobleImg, binartEdgeImg, 0, 255, CV_THRESH_OTSU + CV_THRESH_BINARY);

//利用closing結合邊緣試著將多邊緣區域結合成一圖塊(blob) cv::Mat element = cv::getStructuringElement(cv::MORPH_RECT, cv::Size(17, 3)); cv::morphologyEx(binartEdgeImg, binartEdgeImg, CV_MOP_CLOSE, element);
//找連結的圖塊的輪廓
std::vector< std::vector< cv::Point> > contours;
findContours(binartEdgeImg, contours, CV_RETR_EXTERNAL, CV_CHAIN_APPROX_NONE);

//開始處理偵測到的輪廓
std::vector >::iterator itc = contours.begin();
std::vector rects;

//Remove patch that are no inside limits of aspect ratio and area.
while (itc != contours.end())
{
   //Create bounding rect of object
   cv::RotatedRect mAr = cv::minAreaRect(cv::Mat(*itc));
   if (!verifyCarPlateSizes(mAr))
   {
      itc = contours.erase(itc);
   }
   else
   {
      ++itc;
      rects.push_back(mAr);
   }
}

cv::Mat result;
carImgColor.copyTo(result);
cv::drawContours(result, contours,
-1, // draw all contours
cv::Scalar(255, 0, 0), // in blue
1); // with a thickness of 1

for (int i = 0; i< rects.size(); i++)
{

   //為了找到更適合的車牌區間，利用洪水法的方法來填充相似的色塊，以及配合contour找出的圖塊邊緣來判斷
   cv::circle(result, rects[i].center, 3, cv::Scalar(0, 255, 0), -1);
   //get the min size between width and height
   float minSize = (rects[i].size.width < rects[i].size.height) ? rects[i].size.width : rects[i].size.height;
   minSize = minSize - minSize*0.5;    cv::Mat mask;
   mask.create(carImgColor.rows + 2, carImgColor.cols + 2, CV_8UC1);
   mask = cv::Scalar::all(0);
   int loDiff = 30;
   int upDiff = 30;
   int connectivity = 4;
   int newMaskVal = 255;
   int NumSeeds = 10;
   cv::Rect ccomp;
   int flags = connectivity + (newMaskVal << 8) + CV_FLOODFILL_FIXED_RANGE + CV_FLOODFILL_MASK_ONLY;
   for (int j = 0; j    {
      cv::Point seed;
      seed.x = rects[i].center.x + rand() % (int)minSize - (minSize / 2);
      seed.y = rects[i].center.y + rand() % (int)minSize - (minSize / 2);
      cv::circle(result, seed, 1, cv::Scalar(0, 255, 255), -1);
      int area = floodFill(carImgColor, mask, seed, cv::Scalar(255, 0, 0), &ccomp, cv::Scalar(loDiff, loDiff, loDiff), cv::Scalar(upDiff, upDiff, upDiff), connectivity | (newMaskVal << 8) | CV_FLOODFILL_FIXED_RANGE | CV_FLOODFILL_MASK_ONLY);
   }

   cv::imshow("MASK", mask);
   cv::waitKey(0);

   //Check new floodfill mask match for a correct patch.
   //Get all points detected for get Minimal rotated Rect
   std::vector pointsInterest;
   cv::Mat_::iterator itMask = mask.begin();
   cv::Mat_::iterator end = mask.end();
   for (; itMask != end; ++itMask)
      if (*itMask == 255)
         pointsInterest.push_back(itMask.pos());

   cv::RotatedRect minRect = minAreaRect(pointsInterest);

   if (verifyCarPlateSizes(minRect))
   {
      // rotated rectangle drawing
      cv::Point2f rect_points[4]; minRect.points(rect_points);
      for (int j = 0; j < 4; j++)
         line(result, rect_points[j], rect_points[(j + 1) % 4], cv::Scalar(0, 0, 255), 1, 8);

      //Get rotation matrix
      float r = (float)minRect.size.width / (float)minRect.size.height;
      float angle = minRect.angle;
      if (r<1)
         angle = 90 + angle;
      cv::Mat rotmat = getRotationMatrix2D(minRect.center, angle, 1);

      //Create and rotate image
      cv::Mat img_rotated;
      cv::warpAffine(carImgColor, img_rotated, rotmat, carImgColor.size(), CV_INTER_CUBIC);

      //Crop image
      cv::Size rect_size = minRect.size;
      if (r < 1)
         std::swap(rect_size.width, rect_size.height);
      cv::Mat img_crop;
      getRectSubPix(img_rotated, rect_size, minRect.center, img_crop);

      cv::Mat resultResized;
      resultResized.create(33, 110, CV_8UC3);
      resize(img_crop, resultResized, resultResized.size(), 0, 0, cv::INTER_CUBIC);
      //Equalize croped image
      cv::Mat grayResult;
      cv::cvtColor(resultResized, grayResult, CV_BGR2GRAY);
      cv::blur(grayResult, grayResult, cv::Size(3, 3));
      grayResult = histeq(grayResult);

      std::string filename = "file";
      std::stringstream numS;
      numS << i;
      filename = filename + numS.str() + ".bmp";
      cv::imshow("carPlate", grayResult);
      cv::waitKey(0);
      cv::imwrite(filename, grayResult);
      num++;
   }
}

}

int main(int argc, char** argv)
{
cv::Mat carImgInColor = cv::imread("car01.jpg", 1);

cv::Mat carImgIn = cv::imread("car01.jpg", 0);

cv::Mat carImg, carImgColor;

cv::resize(carImgIn, carImg, cv::Size(800,600));
cv::resize(carImgInColor, carImgColor, cv::Size(800, 600));

int num = 0;

carPlateSegment(carImg, carImgColor, num);
return 0;
}