openCV 形态学 腐蚀、膨胀、开操作和比操作、形态梯度 、顶帽、黑帽



一、opencv几个形态学函数定义

形态学函数的头文件：#include <opencv2/imgproc/imgproc.hpp>
函数定义如下：

[cpp]
view plaincopyprint?

//! erodes the image (applies the local minimum operator)
CV_EXPORTS_W void erode( InputArray src, OutputArray dst, InputArray kernel,
Point anchor=Point(-1,-1), int iterations=1,
int borderType=BORDER_CONSTANT,
const Scalar& borderValue=morphologyDefaultBorderValue() );

//! dilates the image (applies the local maximum operator)
CV_EXPORTS_W void dilate( InputArray src, OutputArray dst, InputArray kernel,
Point anchor=Point(-1,-1), int iterations=1,
int borderType=BORDER_CONSTANT,
const Scalar& borderValue=morphologyDefaultBorderValue() );

//! applies an advanced morphological operation to the image
CV_EXPORTS_W void morphologyEx( InputArray src, OutputArray dst,
int op, InputArray kernel,
Point anchor=Point(-1,-1), int iterations=1,
int borderType=BORDER_CONSTANT,
const Scalar& borderValue=morphologyDefaultBorderValue() );

//! erodes the image (applies the local minimum operator)
CV_EXPORTS_W void erode( InputArray src, OutputArray dst, InputArray kernel,
                         Point anchor=Point(-1,-1), int iterations=1,
                         int borderType=BORDER_CONSTANT,
                         const Scalar& borderValue=morphologyDefaultBorderValue() );

//! dilates the image (applies the local maximum operator)
CV_EXPORTS_W void dilate( InputArray src, OutputArray dst, InputArray kernel,
                          Point anchor=Point(-1,-1), int iterations=1,
                          int borderType=BORDER_CONSTANT,
                          const Scalar& borderValue=morphologyDefaultBorderValue() );

//! applies an advanced morphological operation to the image
CV_EXPORTS_W void morphologyEx( InputArray src, OutputArray dst,
                                int op, InputArray kernel,
                                Point anchor=Point(-1,-1), int iterations=1,
                                int borderType=BORDER_CONSTANT,
                                const Scalar& borderValue=morphologyDefaultBorderValue() );

开运算 (Opening)

开运算是通过先对图像腐蚀再膨胀实现的。



能够排除小团块物体(假设物体较背景明亮)

请看下面，左图是原图像，右图是采用开运算转换之后的结果图。 观察发现字母拐弯处的白色空间消失。

闭运算(Closing)

闭运算是通过先对图像膨胀再腐蚀实现的。



能够排除小型黑洞(黑色区域)。

形态梯度(Morphological Gradient)

膨胀图与腐蚀图之差



能够保留物体的边缘轮廓，如下所示:

顶帽(Top Hat)

原图像与开运算结果图之差

黑帽(Black Hat)

闭运算结果图与原图像之差

二、形态学函数解析

erode（腐蚀函数）： InputArray src, 原图像
OutputArray dst, 结果输出图像
InputArray kernel, 结构元素

Point anchor=Point(-1,-1), 结构元素的原点
int iterations=1, 迭代次数

dilate（膨胀函数）：InputArray src, 原图像
OutputArray dst, 结果输出图像
InputArray kernel, 结构元素

Point anchor=Point(-1,-1), 结构元素的原点
int iterations=1, 迭代次数
morphologyEx（形态学函数）：InputArray src, 原图像
OutputArray dst, 结果输出图像
int op，cv::MORPH_OPEN（打开） cv::MORPH_CLOSE（关闭）
Gradient: MORPH_GRADIENT（梯度）

Top Hat: MORPH_TOPHAT（顶帽） Black Hat: MORPH_BLACKHAT（黑帽）


InputArray kernel, 结构元素

Point anchor=Point(-1,-1), 结构元素的原点
int iterations=1, 迭代次数

三、形态学函数使用举例及结果分析

程序对源图像进行多种形态学处理，不同参数的膨胀（如函数默认的结构元素是3x3，改为7x7，还可以设置多次迭代），腐蚀，打开，关闭，以及开关的级联。


[cpp]
view plaincopyprint?

#include <opencv2/core/core.hpp>
#include <opencv2/imgproc/imgproc.hpp>
#include <opencv2/highgui/highgui.hpp>

int main()
{
// Read input image
cv::Mat image= cv::imread("binary.bmp");
if (!image.data)
return 0;

// Display the image
cv::namedWindow("Image");
cv::imshow("Image",image);

// Erode the image
cv::Mat eroded;
cv::erode(image,eroded,cv::Mat());

// Display the eroded image
cv::namedWindow("Eroded Image");
cv::imshow("Eroded Image",eroded);
cv::imwrite("eroded.bmp",eroded);

// Dilate the image
cv::Mat dilated;
cv::dilate(image,dilated,cv::Mat());

// Display the dialted image
cv::namedWindow("Dilated Image");
cv::imshow("Dilated Image",dilated);
cv::imwrite("dilated.bmp",dilated);

// Erode the image with a larger s.e.
cv::Mat element(7,7,CV_8U,cv::Scalar(1));
cv::erode(image,eroded,element);

// Display the eroded image
cv::namedWindow("Eroded Image (7x7)");
cv::imshow("Eroded Image (7x7)",eroded);
cv::imwrite("Eroded Image 7x7.bmp",eroded);

// Erode the image 3 times.
cv::erode(image,eroded,cv::Mat(),cv::Point(-1,-1),3);

// Display the eroded image
cv::namedWindow("Eroded Image (3 times)");
cv::imshow("Eroded Image (3 times)",eroded);
cv::imwrite("Eroded Image 3 times.bmp",eroded);

// Close the image
cv::Mat element5(5,5,CV_8U,cv::Scalar(1));
cv::Mat closed;
cv::morphologyEx(image,closed,cv::MORPH_CLOSE,element5);

// Display the opened image
cv::namedWindow("Closed Image");
cv::imshow("Closed Image",closed);
cv::imwrite("Closed Image.bmp",closed);

// Open the image
cv::Mat opened;
cv::morphologyEx(image,opened,cv::MORPH_OPEN,element5);

// Display the opened image
cv::namedWindow("Opened Image");
cv::imshow("Opened Image",opened);
cv::imwrite("Opened Image.bmp",opened);

// Close and Open the image
cv::morphologyEx(image,image,cv::MORPH_CLOSE,element5);
cv::morphologyEx(image,image,cv::MORPH_OPEN,element5);

// Display the close/opened image
cv::namedWindow("Closed and Opened Image");
cv::imshow("Closed and Opened Image",image);
cv::imwrite("Closed and Opened Image.bmp",image);

// Read input image
image= cv::imread("binary.bmp");

// Open and Close the image
cv::morphologyEx(image,image,cv::MORPH_OPEN,element5);
cv::morphologyEx(image,image,cv::MORPH_CLOSE,element5);

// Display the close/opened image
cv::namedWindow("Opened and Closed Image");
cv::imshow("Opened and Closed Image",image);
cv::imwrite("Opened and Closed Image.bmp",image);

cv::waitKey();
return 0;
}

#include <opencv2/core/core.hpp>
#include <opencv2/imgproc/imgproc.hpp>
#include <opencv2/highgui/highgui.hpp>

int main()
{
	// Read input image
	cv::Mat image= cv::imread("binary.bmp");
	if (!image.data)
		return 0; 

    // Display the image
	cv::namedWindow("Image");
	cv::imshow("Image",image);

	// Erode the image
	cv::Mat eroded;
	cv::erode(image,eroded,cv::Mat());

    // Display the eroded image
	cv::namedWindow("Eroded Image");
	cv::imshow("Eroded Image",eroded);
	cv::imwrite("eroded.bmp",eroded);

	// Dilate the image
	cv::Mat dilated;
	cv::dilate(image,dilated,cv::Mat());

    // Display the dialted image
	cv::namedWindow("Dilated Image");
	cv::imshow("Dilated Image",dilated);
	cv::imwrite("dilated.bmp",dilated);

	// Erode the image with a larger s.e.
	cv::Mat element(7,7,CV_8U,cv::Scalar(1));
	cv::erode(image,eroded,element);

    // Display the eroded image
	cv::namedWindow("Eroded Image (7x7)");
	cv::imshow("Eroded Image (7x7)",eroded);
	cv::imwrite("Eroded Image 7x7.bmp",eroded);

	// Erode the image 3 times.
	cv::erode(image,eroded,cv::Mat(),cv::Point(-1,-1),3);

    // Display the eroded image
	cv::namedWindow("Eroded Image (3 times)");
	cv::imshow("Eroded Image (3 times)",eroded);
	cv::imwrite("Eroded Image 3 times.bmp",eroded);

	// Close the image
	cv::Mat element5(5,5,CV_8U,cv::Scalar(1));
	cv::Mat closed;
	cv::morphologyEx(image,closed,cv::MORPH_CLOSE,element5);
	
    // Display the opened image
	cv::namedWindow("Closed Image");
	cv::imshow("Closed Image",closed);
	cv::imwrite("Closed Image.bmp",closed);

	// Open the image
	cv::Mat opened;
	cv::morphologyEx(image,opened,cv::MORPH_OPEN,element5);

    // Display the opened image
	cv::namedWindow("Opened Image");
	cv::imshow("Opened Image",opened);
	cv::imwrite("Opened Image.bmp",opened);

	// Close and Open the image
	cv::morphologyEx(image,image,cv::MORPH_CLOSE,element5);
	cv::morphologyEx(image,image,cv::MORPH_OPEN,element5);

    // Display the close/opened image
	cv::namedWindow("Closed and Opened Image");
	cv::imshow("Closed and Opened Image",image);
	cv::imwrite("Closed and Opened Image.bmp",image);

	// Read input image
	image= cv::imread("binary.bmp");

	// Open and Close the image
	cv::morphologyEx(image,image,cv::MORPH_OPEN,element5);
	cv::morphologyEx(image,image,cv::MORPH_CLOSE,element5);

    // Display the close/opened image
	cv::namedWindow("Opened and Closed Image");
	cv::imshow("Opened and Closed Image",image);
	cv::imwrite("Opened and Closed Image.bmp",image);

	cv::waitKey();
	return 0;
}

四、程序结果显示和分析:
binary.bmp(源图像)



eroded.bmp （腐蚀）消除一些小的噪声



dilated.bmp（膨胀）填充物体的空洞，膨胀物体



Eroded Image 7x7.bmp（腐蚀，结构元素尺寸是7x7）相比结构元素为默认3x3的相比，源图像的噪声跟大点的噪声也消失了



Eroded Image 3 times.bmp（腐蚀 3次迭代）迭代后结果不变



Opened Image.bmp（先腐蚀后膨胀）消除小噪声，然后再膨胀，平滑大物体，分离小物体



Closed Image.bmp（关闭 先膨胀在腐蚀）填充细小空洞，连接相邻物体，平滑边界



Opened and Closed Image.bmp 先打开在关闭



Closed and Opened Image.bmp先关闭再打开