【OpenCV】使用floodfill()实现PhotoShop魔棒功能

OpenCV中看到一个很有意思的函数:floodfill()

使用给定颜色填充一个联通的区域

C++: int floodFill(InputOutputArray image, Point seedPoint, 
Scalar newVal, Rect* rect=0, Scalar loDiff=Scalar(), 
Scalar upDiff=Scalar(), int flags=4 )

一个简单的例子:

#include "opencv2/imgproc/imgproc.hpp"
#include "opencv2/highgui/highgui.hpp"

#include <iostream>

using namespace cv;
using namespace std;


//floodfill()
//Fills a connected component with the given color.

static void help()
{
 cout << "\nThis program demonstrated the floodFill() function\n"
  "Call:\n"
  "./ffilldemo [image_name -- Default: fruits.jpg]\n" << endl;

cout << "Hot keys: \n"
  "\tESC - quit the program\n"
  "\tc - switch color/grayscale mode\n"
  "\tm - switch mask mode\n"
  "\tr - restore the original image\n"
  "\ts - use null-range floodfill\n"
  "\tf - use gradient floodfill with fixed(absolute) range\n"
  "\tg - use gradient floodfill with floating(relative) range\n"
  "\t4 - use 4-connectivity mode\n"
  "\t8 - use 8-connectivity mode\n" << endl;
}

Mat image0, image, gray, mask;
int ffillMode = 1;
int loDiff = 20, upDiff = 20;
int connectivity = 4;
int isColor = true;
bool useMask = false;
int newMaskVal = 255;

static void onMouse( int event, int x, int y, int, void* )
{
 if( event != CV_EVENT_LBUTTONDOWN )
  return;

Point seed = Point(x,y);
 int lo = ffillMode == 0 ? 0 : loDiff;
 int up = ffillMode == 0 ? 0 : upDiff;
 int flags = connectivity + (newMaskVal << 8) +
  (ffillMode == 1 ? CV_FLOODFILL_FIXED_RANGE : 0);
 int b = (unsigned)theRNG() & 255;
 int g = (unsigned)theRNG() & 255;
 int r = (unsigned)theRNG() & 255;
 Rect ccomp;

Scalar newVal = isColor ? Scalar(b, g, r) : Scalar(r*0.299 + g*0.587 + b*0.114);
 Mat dst = isColor ? image : gray;
 int area;

if( useMask )
 {
  threshold(mask, mask, 1, 128, CV_THRESH_BINARY);
  area = floodFill(dst, mask, seed, newVal, &ccomp, Scalar(lo, lo, lo),
   Scalar(up, up, up), flags);
  imshow( "mask", mask );
 }
 else
 {
  area = floodFill(dst, seed, newVal, &ccomp, Scalar(lo, lo, lo),
   Scalar(up, up, up), flags);
 }

imshow("image", dst);
 cout << area << " pixels were repainted\n";
}


int main( )
{
 char* filename="0.png";
 image0 = imread(filename, 1);

if( image0.empty() )
 {
  cout << "Image empty. Usage: ffilldemo <image_name>\n";
  return 0;
 }
 help();
 image0.copyTo(image);
 cvtColor(image0, gray, CV_BGR2GRAY);
 mask.create(image0.rows+2, image0.cols+2, CV_8UC1);

namedWindow( "image", 0 );
 createTrackbar( "lo_diff", "image", &loDiff, 255, 0 );
 createTrackbar( "up_diff", "image", &upDiff, 255, 0 );

setMouseCallback( "image", onMouse, 0 );

for(;;)
 {
  imshow("image", isColor ? image : gray);

int c = waitKey(0);
  if( (c & 255) == 27 )
  {
   cout << "Exiting ...\n";
   break;
  }
  switch( (char)c )
  {
  case 'c':
   if( isColor )
   {
    cout << "Grayscale mode is set\n";
    cvtColor(image0, gray, CV_BGR2GRAY);
    mask = Scalar::all(0);
    isColor = false;
   }
   else
   {
    cout << "Color mode is set\n";
    image0.copyTo(image);
    mask = Scalar::all(0);
    isColor = true;
   }
   break;
  case 'm':
   if( useMask )
   {
    destroyWindow( "mask" );
    useMask = false;
   }
   else
   {
    namedWindow( "mask", 0 );
    mask = Scalar::all(0);
    imshow("mask", mask);
    useMask = true;
   }
   break;
  case 'r':
   cout << "Original image is restored\n";
   image0.copyTo(image);
   cvtColor(image, gray, CV_BGR2GRAY);
   mask = Scalar::all(0);
   break;
  case 's':
   cout << "Simple floodfill mode is set\n";
   ffillMode = 0;
   break;
  case 'f':
   cout << "Fixed Range floodfill mode is set\n";
   ffillMode = 1;
   break;
  case 'g':
   cout << "Gradient (floating range) floodfill mode is set\n";
   ffillMode = 2;
   break;
  case '4':
   cout << "4-connectivity mode is set\n";
   connectivity = 4;
   break;
  case '8':
   cout << "8-connectivity mode is set\n";
   connectivity = 8;
   break;
  }
 }

return 0;
}

点击图标改变图像中的连图区域的颜色: 

【OpenCV】使用floodfill()实现PhotoShop魔棒功能

--------------------------------------分割线 --------------------------------------

内容版权声明:除非注明,否则皆为本站原创文章。

转载注明出处:https://www.heiqu.com/f4e586ab6d501d0a0e0b89be74ee291c.html