1. 程式人生 > >基於OpenCV 的影象分割

基於OpenCV 的影象分割

1、影象閾值化
原始碼:

#include "opencv2/highgui/highgui.hpp"
#include "opencv2/imgproc/imgproc.hpp"
#include <iostream>
using namespace std;
using namespace cv;
int thresholds=50;
int model=2;
Mat image,srcimage;
void track(int ,void *)
{
    Mat result;
    threshold(srcimage,result,thresholds,255,CV_THRESH_BINARY);
    //imshow("原圖",result);
	if(model==0)
	{
		threshold(srcimage,result,thresholds,255,CV_THRESH_BINARY);
		imshow("分割",result);
	}
	if(model==1)
	{
		threshold(srcimage,result,thresholds,255,THRESH_BINARY_INV);
		imshow("分割",result);	
	}
	if(model==2)
	{
		threshold(srcimage,result,thresholds,255,THRESH_TRUNC);
		imshow("分割",result);
	}
	if(model==3)
	{
		threshold(srcimage,result,thresholds,255,THRESH_TOZERO);
		imshow("分割",result);
	}
	if(model==4)
	{
		threshold(srcimage,result,thresholds,255,THRESH_TOZERO_INV);
		imshow("分割",result);
	}
}
int main()
{
    image=imread("2.2.tif");
    if(!image.data)
    {
        return 0;
    }
    cvtColor(image,srcimage,CV_BGR2GRAY);
    namedWindow("分割",WINDOW_AUTOSIZE);
    cv::createTrackbar("閾a值:","分割",&thresholds,255,track);
	cv::createTrackbar("模式:","分割",&model,4,track);
    track(thresholds,0);
	track(model,0);
    waitKey(0);
    return 0;
}

實現結果:
在這裡插入圖片描述
2、閾值處理

    //閾值處理
    #include "opencv2/core/core.hpp"
    #include "opencv2/highgui/highgui.hpp"
    #include "opencv2/imgproc/imgproc.hpp"
    
    using namespace cv;
    using namespace std;
    
    int main()
    {
    	printf("鍵盤按鍵ESC--退出程式");
    	Mat g_srcImage = imread("1.tif",0);
    	if(!g_srcImage.data)
    	{
    		printf("讀取圖片失敗");
    	}
    	imshow("原始圖",g_srcImage);
    
    	//大津法閾值分割顯示
    	/*大津法,簡稱OTSU.它是按影象的灰度特性,將影象分成背景
    	和目標2部分。背景和目標之間的類間方差越大,說明構成影象
    	的2部分的差別越大,當部分目標錯分為背景或部分背景錯分為
    	目標都會導致2部分差別變小。*/
    	Mat OtsuImage;
    	threshold(g_srcImage,OtsuImage,0,255,THRESH_OTSU);//0不起作用,可為任意閾值
    	imshow("OtsuImage",OtsuImage);
    
    	//自適應分割並顯示
    	Mat AdaptImage;
    	//THRESH_BINARY_INV:引數二值化取反
    	adaptiveThreshold(g_srcImage,AdaptImage,255,0,THRESH_BINARY_INV,7,8);
    	imshow("AdaptImage",AdaptImage);
    
    	while(1)
    	{
    		int key;
    		key = waitKey(20);
    		if((char)key == 27)
    		{ break; }
    	}
    }

在這裡插入圖片描述
3、拉普拉斯檢測

//Laplacian檢測
#include "opencv2/core/core.hpp"
#include "opencv2/highgui/highgui.hpp"
#include "opencv2/imgproc/imgproc.hpp"
using namespace cv;
using namespace std;

/*,在只關心邊緣的位置而不考慮其周圍的象素灰度差值時比較合適。
Laplace 運算元對孤立象素的響應要比對邊緣或線的響應要更強烈,因此
只適用於無噪聲圖象。存在噪聲情況下,使用 Laplacian 運算元檢測邊
緣之前需要先進行低通濾波。*/
int main()
{
	Mat src,src_gray,dst,abs_dst;
	src = imread("1.jpg");
	imshow("原始影象",src);

	//高斯濾波
	GaussianBlur(src,src,Size(3,3),0,0,BORDER_DEFAULT);
	//轉化為灰度圖,輸入只能為單通道
	cvtColor(src,src_gray,CV_BGR2GRAY);

	Laplacian(src_gray,dst,CV_16S,3,1,0,BORDER_DEFAULT);
	convertScaleAbs(dst,abs_dst);
	imshow("效果圖Laplace變換",abs_dst);
	waitKey();
	return 0;

}

效果圖:
在這裡插入圖片描述
4、canny演算法的邊緣檢測
原始碼

#include "opencv2/core/core.hpp"
#include "opencv2/highgui/highgui.hpp"
#include "opencv2/imgproc/imgproc.hpp"
using namespace cv;
using namespace std;
/*如果某一畫素位置的幅值超過高閾值,該畫素被保留為邊緣畫素。如果某
一畫素位置的幅值小於低閾值,該畫素被排除。如果某一畫素位置的幅值在
兩個閾值之間,該畫素僅僅在連線到一個高於高閾值的畫素時被保留。 */
int main()
{
	Mat picture2=imread("1.jpg");
	Mat new_picture2;
	Mat picture2_1=picture2.clone();
	Mat gray_picture2 , edge , new_edge;
	imshow("【原始圖】Canny邊緣檢測" , picture2);
	Canny(picture2_1 , new_picture2 ,150 , 100 ,3  );
	imshow("【效果圖】Canny邊緣檢測", new_picture2 );
	Mat dstImage,grayImage;
	//dstImage與srcImage同大小型別
	dstImage.create(picture2_1.size() , picture2_1.type());
	cvtColor(picture2_1,gray_picture2,CV_BGR2GRAY);//轉化為灰度圖
	blur(gray_picture2 , edge , Size(3,3));//用3x3的核心降噪
	Canny(edge,edge,3,9,3);
	dstImage = Scalar::all(0);//將dst內所有元素設定為0
	//使用canny運算元的邊緣圖edge作為掩碼,將原圖拷貝到dst中
	picture2_1.copyTo(dstImage,edge);
	imshow("效果圖Canny邊緣檢測2",dstImage);
	waitKey();
}

在這裡插入圖片描述
5、影象的分水嶺演算法
原始碼:

#include “opencv2/core/core.hpp”
#include “opencv2/highgui/highgui.hpp”
#include “opencv2/imgproc/imgproc.hpp”
#include

using namespace cv;
using namespace std;

#define WINDOW_NAME1 “顯示/操作視窗”
#define WINDOW_NAME2 “分水嶺演算法效果圖”

Mat g_maskImage,g_srcImage;
Point prevPt(-1,-1);

static void ShowHelpText();
static void on_Mouse(int event,int x,int y,int flags,void*);

//輸出一些幫助資訊
static void ShowHelpText()
{
printf(“當前使用的版本為:“CV_VERSION);
printf(”\n”);
printf(“分水嶺演算法—點中圖片進行滑鼠或按鍵操作\n”);
printf(“請先用滑鼠在圖片視窗中標記出大致的區域,\n然後再按鍵【1】或者【space】啟動演算法”);
printf("\n按鍵操作說明:\n"
“鍵盤按鍵【1】或者【space】–執行的分水嶺分割演算法\n”
“鍵盤按鍵【2】–回覆原始圖片\n”
“鍵盤按鍵【ESC】–退出程式\n”);
}

static void on_Mouse(int event,int x,int y,int flags,void*)
{
if(x<0||x>=g_srcImage.cols||y<0||y>=g_srcImage.rows)
return;

if(event == CV_EVENT_LBUTTONUP||!(flags & CV_EVENT_FLAG_LBUTTON))
	prevPt = Point(-1,-1);

else if(event == CV_EVENT_LBUTTONDOWN)
	prevPt= Point(x,y);

else if(event == CV_EVENT_MOUSEMOVE && (flags & CV_EVENT_FLAG_LBUTTON))
{
	Point pt(x,y);
	if(prevPt.x<0)
		prevPt = pt;
	line(g_maskImage,prevPt,pt,Scalar::all(255),5,8,0);
	line(g_srcImage,prevPt,pt,Scalar::all(255),5,8,0);
	prevPt = pt;
	imshow(WINDOW_NAME1,g_srcImage);
}

}

int main(int argc,char** argv)
{
system(“color A5”);

ShowHelpText();

g_srcImage = imread("1.jpg",1);
imshow(WINDOW_NAME1,g_srcImage);
Mat srcImage,grayImage;
g_srcImage.copyTo(srcImage);
cvtColor(g_srcImage,g_maskImage,CV_BGR2GRAY);
cvtColor(g_maskImage,grayImage,CV_GRAY2BGR);//灰度圖轉BGR3通道,但每通道的值都是原先單通道的值,所以也是顯示灰色的
g_maskImage = Scalar::all(0);//黑

setMouseCallback(WINDOW_NAME1,on_Mouse,0);

while(1)
{
	int c = waitKey(0);
	if((char)c == 27)
		break;
	if((char)c == '2')
	{
		g_maskImage = Scalar::all(0);//黑
		srcImage.copyTo(g_srcImage);
		imshow("image",g_srcImage);
	}
	if((char)c == '1'||(char)c == ' ')
	{
		int i,j,compCount = 0;
		vector<vector<Point>> contours;//定義輪廓
		vector<Vec4i> hierarchy;//定義輪廓的層次

		findContours(g_maskImage,contours,hierarchy,RETR_CCOMP,CHAIN_APPROX_SIMPLE);
		if(contours.empty())
			continue;
		Mat maskImage(g_maskImage.size(),CV_32S);
		maskImage = Scalar::all(0);

		for(int index = 0;index >= 0;index = hierarchy[index][0],compCount++)
			drawContours(maskImage,contours,index,Scalar::all(compCount+1),-1,8,hierarchy,INT_MAX);
		if(compCount == 0)
			continue;
		vector<Vec3b> colorTab;
		for(i=0;i<compCount;i++)
		{
			int b = theRNG().uniform(0,255);
			int g = theRNG().uniform(0,255);
			int r = theRNG().uniform(0,255);
			colorTab.push_back(Vec3b((uchar)b,(uchar)g,(uchar)r));
		}
			//計算處理時間並輸出到視窗中
		double dTime = (double)getTickCount();
		watershed(srcImage,maskImage);
		dTime = (double)getTickCount()-dTime;
		printf("\t處理時間=%gms\n",dTime*1000./getTickFrequency());
		//雙層迴圈,將分水嶺影象遍歷存入watershedImage中
		Mat watershedImage(maskImage.size(),CV_8UC3);
		for(i=0;i<maskImage.rows;i++)
			for(j=0;j<maskImage.cols;j++)
			{
				int index = maskImage.at<int>(i,j);
				if(index == -1)
					watershedImage.at<Vec3b>(i,j) = Vec3b(255,255,255);
				else if(index<=0||index>compCount)
					watershedImage.at<Vec3b>(i,j) = Vec3b(0,0,0);
				else
					watershedImage.at<Vec3b>(i,j) = colorTab[index-1];	
			}
			//混合灰度圖和分水嶺效果圖並顯示最終的視窗
			watershedImage = watershedImage*0.5+grayImage*0.5;
			imshow(WINDOW_NAME2,watershedImage);								
	}	
}
waitKey();
return 0;

}


效果:
在這裡插入圖片描述
![在這裡插入圖片描述](https://img-blog.csdnimg.cn/20181121123847750.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3ZpY3RvX2NoYW8=,size_16,color_FFFFFF,t_70)