基于OpenCV与JVM实现矩阵处理图像基于OpenCV与JVM实现矩阵处理图

图片裁剪
图片模糊处理
子矩阵生成矩阵
从图片子矩阵生成矩阵

submat(int rowStart, int rowEnd, int colStart, int colEnd) 函数的返回值是一个矩阵对象。内容是原图的子矩阵或子区域。

文章图片

首先我们用imread来读取图片，然后输出矩阵对象本身的一些信息

import org.opencv.core.CvType; import org.opencv.core.Mat; import org.opencv.core.Core; import org.opencv.core.MatOfInt; import org.opencv.imgcodecs.Imgcodecs; import origami.Origami; public class HelloCv {public static void main(String[] args) throws Exception {System.loadLibrary(Core.NATIVE_LIBRARY_NAME); Mat mat = Imgcodecs.imread("./images/test.jpg",Imgcodecs.IMREAD_GRAYSCALE); System.out.println(mat); }}

文章图片

由于这个矩阵是原始图片，所以它的isSubmat是false。
现在我们使用submat函数的第一种形式，输入参数是每一行和每一列的起始和终止值。

图片裁剪

import org.opencv.core.CvType; import org.opencv.core.Mat; import org.opencv.core.Core; import org.opencv.core.MatOfInt; import org.opencv.imgcodecs.Imgcodecs; import origami.Origami; public class HelloCv {public static void main(String[] args) throws Exception {System.loadLibrary(Core.NATIVE_LIBRARY_NAME); Mat mat = Imgcodecs.imread("C:/HWKJ/ZRQ/OpenCv/matrixcv/images/test.jpg"); System.out.println(mat); Mat submat = mat.submat(200, 240, 300, 350); System.out.println(submat); }}

文章图片

这里注意submat里的尺寸，尺寸根据原图的尺寸,超出原图的尺寸会报错，报错如下

文章图片

然后我们输出裁剪的图片。
那么如何确认你想要截取图片的区域范围呢?也就是说怎么确定这四个参数的填写？我们以下图为例

文章图片

截取后的图片

文章图片

另外两种submat方式
Range?(int row,int column)
row：宽开始结束范围
column：高开始结束范围

Mat submat2 = mat.submat(new Range(20,300),new Range(100,500)); Imgcodecs.imwrite("./images/output2.png",submat2);

Rect?(int x, int y,int width, int height)
x：横坐标
【基于OpenCV与JVM实现矩阵处理图像】y：纵坐标
width ：宽
height：高

Mat submat3 = mat.submat(new Rect(0,200,100,100)); //submat3.setTo(new Scalar(255,0,0)); //将图片绘制为蓝色Imgcodecs.imwrite("./images/output3.png",submat3);

文章图片

打开setTo如下：

文章图片

Imgcodecs.imwrite("./images/blurtest.png",mat);

文章图片

完整代码：

import org.opencv.core.CvType; import org.opencv.core.Scalar; import org.opencv.core.Mat; import org.opencv.core.Rect; import org.opencv.core.Range; import org.opencv.core.Core; import org.opencv.core.Size; import org.opencv.core.MatOfInt; import org.opencv.imgcodecs.Imgcodecs; import org.opencv.imgproc.Imgproc; import origami.Origami; public class HelloCv {public static void main(String[] args) throws Exception {System.loadLibrary(Core.NATIVE_LIBRARY_NAME); Mat mat = Imgcodecs.imread("C:/HWKJ/ZRQ/OpenCv/matrixcv/images/test.jpg"); System.out.println(mat); Mat submat = mat.submat(200, 400, 200, 550); //System.out.println(submat); Imgcodecs.imwrite("./images/output.png",submat); Mat submat2 = mat.submat(new Range(20,300),new Range(100,500)); Imgcodecs.imwrite("./images/output2.png",submat2); Mat submat3 = mat.submat(new Rect(0,200,400,200)); submat3.setTo(new Scalar(255,0,0)); Imgcodecs.imwrite("./images/output3.png",submat3); //Imgproc.blur(submat,submat,new Size(25.0,25.0)); Imgcodecs.imwrite("./images/blurtest.png",mat); }}

图片模糊处理

import org.opencv.core.CvType; import org.opencv.core.Mat; import org.opencv.core.Core; import org.opencv.core.Size; import org.opencv.core.MatOfInt; import org.opencv.imgcodecs.Imgcodecs; import org.opencv.imgproc.Imgproc; import origami.Origami; public class HelloCv {public static void main(String[] args) throws Exception {System.loadLibrary(Core.NATIVE_LIBRARY_NAME); Mat mat = Imgcodecs.imread("C:/HWKJ/ZRQ/OpenCv/matrixcv/images/test.jpg"); System.out.println(mat); Mat submat = mat.submat(200, 400, 200, 550); //System.out.println(submat); //Imgcodecs.imwrite("./images/output.png",submat); Imgproc.blur(submat,submat,new Size(25.0,25.0)); System.out.println("after:"+mat); Imgcodecs.imwrite("./images/blurtest.png",mat); }}

文章图片

子矩阵生成矩阵 setTo和copyTo是OpenCv中两个非常重要的函数。
setTo可以将一个矩阵中的所有像素设置为指定的颜色
copyTo可以将一个已有的矩阵复制到另一个矩阵之中。
第一个颜色值代表蓝色的深度，第二个值代表绿色的深度，最后一个值代表红色的深度。

//获取红绿蓝Scalar Red = new Scalar(0,0,255); Scalar Green = new Scalar(0,255,0); Scalar Blue = new Scalar(255,0,0);

我们把这些颜色当作RGB的补充色。因此把其他通道设置为最大值255，主通道设置为0。蓝绿色是红色的补充色，所以红色值通道被设为0，而另外两个通道为255；
定义蓝绿色、品红色和黄色

Scalar cyan = new Scalar(255,255,0); Scalarmagena= new Scalar(255,0,255); Scalar yellow = new Scalar(0,255,255);

下面我们使用setTo将子矩阵设置为给定的Scalar颜色

private void setColors(Mat mat ,boolean comp,int row){for (int i = 0; i <3 ; i++) {Mat sub = mat.submat(row*100,row*100+100,i*100,i*100+100); if(comp){//RGBif (i==0){sub.setTo(Red); }if (i==1){sub.setTo(Green); }if (i==2){sub.setTo(Blue); }}else {//cmyif (i==0){sub.setTo(cyan); }if (i==1){sub.setTo(magena); }if (i==2){sub.setTo(yellow); }}}}

接下来，我们创建一个包含三个颜色通道矩阵，并且填充它的第一行和第二行
完整代码：

import org.opencv.core.CvType; import org.opencv.core.Scalar; import org.opencv.core.Mat; import org.opencv.core.Rect; import org.opencv.core.Range; import org.opencv.core.Core; import org.opencv.core.Size; import org.opencv.core.MatOfInt; import org.opencv.imgcodecs.Imgcodecs; import org.opencv.imgproc.Imgproc; import origami.Origami; public class HelloCv1 {public staticScalar Red = new Scalar(0,0,255); public staticScalar Green = new Scalar(0,255,0); public staticScalar Blue = new Scalar(255,0,0); public staticScalar cyan = new Scalar(255,255,0); public staticScalarmagena= new Scalar(255,0,255); public staticScalar yellow = new Scalar(0,255,255); public static void main(String[] args) throws Exception {System.loadLibrary(Core.NATIVE_LIBRARY_NAME); Mat mat = new Mat(200,300,CvType.CV_8UC3); setColors(mat,false,1); setColors(mat,true,0); Imgcodecs.imwrite("./images/rgbcmy.png",mat); }static void setColors(Mat mat ,boolean comp,int row){for (int i = 0; i <3 ; i++) {Mat submat = mat.submat(row*100,row*100+100,i*100,i*100+100); if(comp){//RGBif (i==0){submat.setTo(Red); }if (i==1){submat.setTo(Green); }if (i==2){submat.setTo(Blue); }}else {//cmyif (i==0){submat.setTo(cyan); }if (i==1){submat.setTo(magena); }if (i==2){submat.setTo(yellow); }}}}}

文章图片

从图片子矩阵生成矩阵首先创建一个大小为200x200的矩阵和子矩阵：一个是主矩阵的上部，一个是主矩阵的下部

int width = 200,height = 200; Mat mat1 = new Mat(height,width,CvType.CV_8UC3); Mat top = mat.submat(0,height/2,0,width); Mat bottom = mat.submat(height/2,height,0,width);

然后加载一个图片以创建另一个小矩阵，并把它的大小调整为上部（或下部）的子矩阵大小。这里会引入Imgproc类中的resize函数。
完整代码：

import org.opencv.core.CvType; import org.opencv.core.Scalar; import org.opencv.core.Mat; import org.opencv.core.Rect; import org.opencv.core.Range; import org.opencv.core.Core; import org.opencv.core.Size; import org.opencv.core.MatOfInt; import org.opencv.imgcodecs.Imgcodecs; import org.opencv.imgproc.Imgproc; import origami.Origami; public class HelloCv1 {public staticScalar Red = new Scalar(0,0,255); public staticScalar Green = new Scalar(0,255,0); public staticScalar Blue = new Scalar(255,0,0); public staticScalar cyan = new Scalar(255,255,0); public staticScalarmagena= new Scalar(255,0,255); public staticScalar yellow = new Scalar(0,255,255); public static void main(String[] args) throws Exception {System.loadLibrary(Core.NATIVE_LIBRARY_NAME); int width = 200,height = 300; Mat mat1 = new Mat(height,width,CvType.CV_8UC3); Mat top = mat1.submat(0,height/2,0,width); Mat bottom = mat1.submat(height/2,height,0,width); Mat small = Imgcodecs.imread("./images/test.jpg"); Imgproc.resize(small,small,top.size()); small.copyTo(top); small.copyTo(bottom); Imgcodecs.imwrite("./images/matofpictures.png",mat1); }

注意：设置大小的步骤很关键。复制能够成功，是因为小矩阵和子矩阵的大小是完全相同的，因此复制的时候没有出现任何问题