OpenCV和Java，Photo类图片修复示例

ASR/CV/ML OpenCV和Java，Photo类图片修复示例

Photo提供了降噪Denosing、动态成像HDR、图片拼接等api，类似于Photoshop。

图片填充：

去色演示：

降噪效果演示：

两张噪点图片：

lena1

lena2

import java.awt.event.ItemEvent;
import java.awt.event.ItemListener;
import java.awt.image.BufferedImage;
import java.util.ArrayList;
import java.util.List;
import javax.swing.ButtonGroup;
import javax.swing.ImageIcon;
import javax.swing.JFrame;
import javax.swing.JLabel;
import javax.swing.JRadioButton;
import javax.swing.JSlider;
import javax.swing.event.ChangeEvent;
import javax.swing.event.ChangeListener;
import org.opencv.core.CvType;
import org.opencv.core.Mat;
import org.opencv.imgcodecs.Imgcodecs;
import org.opencv.photo.Photo;

public class InPaint extends JFrame {

    private final JLabel imageView;
    private Mat srcMat;

    public InPaint() {
        srcMat = Imgcodecs.imread("./lena_dirty.jpg");

        JLabel inpaintRadiusLabel = new JLabel("半径:" + 0);
        inpaintRadiusLabel.setBounds(15, 10, 60, 15);
        JSlider inpaintRadiusBar = new JSlider(0, 50);
        inpaintRadiusBar.setValue(0);
        inpaintRadiusBar.setBounds(80, 5, 120, 25);
        JRadioButton nsBtn = new JRadioButton("INPAINT_NS");
        nsBtn.setSelected(true);
        nsBtn.setBounds(230, 10, 120, 15);
        JRadioButton teleaBtn = new JRadioButton("INPAINT_TELEA");
        teleaBtn.setBounds(350, 10, 120, 15);
        ButtonGroup group = new ButtonGroup();
        group.add(nsBtn);
        group.add(teleaBtn);

        inpaintRadiusBar.addChangeListener(new ChangeListener() {
            @Override
            public void stateChanged(ChangeEvent e) {
                double value = inpaintRadiusBar.getValue();
                inpaintRadiusLabel.setText("半径:" + value);
                int type = nsBtn.isSelected() ? 1 : 0;
                inpaint(value, type);
            }
        });

        nsBtn.addItemListener(new ItemListener() {
            @Override
            public void itemStateChanged(ItemEvent e) {
                double value = inpaintRadiusBar.getValue();
                inpaintRadiusLabel.setText("半径:" + value);
                int type = nsBtn.isSelected() ? 1 : 0;
                inpaint(value, type);
            }
        });
        teleaBtn.addItemListener(new ItemListener() {
            @Override
            public void itemStateChanged(ItemEvent e) {
                double value = inpaintRadiusBar.getValue();
                inpaintRadiusLabel.setText("半径:" + value);
                int type = nsBtn.isSelected() ? 1 : 0;
                inpaint(value, type);
            }
        });

        imageView = new JLabel();
        imageView.setBounds(5, 5, 512, 512);
        this.setTitle("图片处理");
        this.setSize(550, 580);
        this.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
        this.getContentPane().setLayout(null);
        this.getContentPane().add(inpaintRadiusLabel);
        this.getContentPane().add(inpaintRadiusBar);
        this.getContentPane().add(nsBtn);
        this.getContentPane().add(teleaBtn);
        this.getContentPane().add(imageView);

        setIcon(srcMat);
    }

    /**
     * 区域填充
     *
     * @param inpaintRadius 数值越大，修复区域越模糊-过渡越好
     * @param flags         本例 INPAINT_TELEA 慢于 INPAINT_NS
     */
    private void inpaint(double inpaintRadius, int flags) {
        Mat dstMat = new Mat(srcMat.size(), CvType.CV_8UC3);
        Mat inpaintMaskMat = Imgcodecs.imread("./lena_dirtymask.jpg", CvType.CV_8UC1); // 模拟手动选择的要修复的区域
        if (1 == flags) {
            Photo.inpaint(srcMat, inpaintMaskMat, dstMat, inpaintRadius, Photo.INPAINT_NS);    // 边缘线算法
        } else {
            Photo.inpaint(srcMat, inpaintMaskMat, dstMat, inpaintRadius, Photo.INPAINT_TELEA); // 临近像素算法
        }
        setIcon(dstMat);
    }
    
    // 去色
    private void decolor() {
        srcMat = Imgcodecs.imread("./lena.jpg");
        Mat grayscaleMat = new Mat();  // 灰阶图，8位单通道
        Mat colorBoostMat = new Mat(); // 彩图，8位三通道
        Photo.decolor(srcMat, grayscaleMat, colorBoostMat);

        setIcon(grayscaleMat);
        setIcon(colorBoostMat);
    }

    // 降噪-去除杂点。图片整体上有种柔化的感觉
    private void fastNlMeansDenoising(){
        Mat dstMat = new Mat();
        
        // 方法1。可用于单通道、3通道图
        // fastNlMeansDenoising(Mat src, Mat& dst, float h = 3, int templateWindowSize = 7, int searchWindowSize = 21)
        Photo.fastNlMeansDenoising(srcMat, dstMat);
        
        // 方法2。按照索引去除杂点。可用于单通道、3通道图
        Mat im1 = Imgcodecs.imread("./lena1.jpg",Imgcodecs.CV_LOAD_IMAGE_COLOR);
        Mat im2 = Imgcodecs.imread("./lena2.jpg",Imgcodecs.CV_LOAD_IMAGE_COLOR);
        List<Mat> srcImgs=new ArrayList<>();
        srcImgs.add(im1); // 杂点图片
        srcImgs.add(im2); // 杂点图片
        srcImgs.add(srcMat);
        // fastNlMeansDenoisingMulti(vector_Mat srcImgs, Mat& dst, int imgToDenoiseIndex, int temporalWindowSize, float h = 3, int templateWindowSize = 7, int searchWindowSize = 21)
        // imgToDenoiseIndex 指srcImgs中"杂点的索引"图片的索引。
        // temporalWindowSize 要去除的杂点周围的像素数量，须奇数。
        Photo.fastNlMeansDenoisingMulti(srcImgs, dstMat, 2, 1);

        // 方法3。可用于3通道图
        // fastNlMeansDenoisingColored(Mat src, Mat& dst, float h = 3, float hColor = 3, int templateWindowSize = 7, int searchWindowSize = 21)
        Photo.fastNlMeansDenoisingColored(srcMat, dstMat);

        // 方法4
        // fastNlMeansDenoisingColoredMulti(vector_Mat srcImgs, Mat& dst, int imgToDenoiseIndex, int temporalWindowSize, float h = 3, float hColor = 3, int templateWindowSize = 7, int searchWindowSize = 21)
        Photo.fastNlMeansDenoisingColoredMulti(srcImgs, dstMat, 2, 1); 

        // 方法5 
        Mat img = Imgcodecs.imread("./lena.jpg",Imgcodecs.CV_LOAD_IMAGE_GRAYSCALE); // 转灰度图 CV_LOAD_IMAGE_GRAYSCALE ：gray = 0.299 * r + 0.587 * g + 0.114 * b ，小数点部分四舍五入为整数
        Mat img1 = Imgcodecs.imread("./lena1.jpg",Imgcodecs.CV_LOAD_IMAGE_GRAYSCALE);
        Mat img2 = Imgcodecs.imread("./lena2.jpg",Imgcodecs.CV_LOAD_IMAGE_GRAYSCALE);
        List<Mat> srcList=new ArrayList<Mat>();
        srcList.add(img);
        // srcList.add(img1);
        // srcList.add(img2); 
        // denoise_TVL1(vector_Mat observations, Mat result, double lambda = 1.0, int niters = 30)
        // lambda 数值越小，图像越模糊-柔化
        // niters 运算次数
        Photo.denoise_TVL1(srcList, dstMat, 10, 10);
        
        setIcon(dstMat);
    }

    private void setIcon(Mat mat) {
        BufferedImage image = CVUtil.matToBufferedImage(mat); // CVUtil 见 http://www.gaohaiyan.com/3229.html
        imageView.setIcon(new ImageIcon(image));
    }
}

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

import java.awt.event.ItemEvent;

import java.awt.event.ItemListener;

import java.awt.image.BufferedImage;

import java.util.ArrayList;

import java.util.List;

import javax.swing.ButtonGroup;

import javax.swing.ImageIcon;

import javax.swing.JFrame;

import javax.swing.JLabel;

import javax.swing.JRadioButton;

import javax.swing.JSlider;

import javax.swing.event.ChangeEvent;

import javax.swing.event.ChangeListener;

import org.opencv.core.CvType;

import org.opencv.core.Mat;

import org.opencv.imgcodecs.Imgcodecs;

import org.opencv.photo.Photo;

public class InPaint extends JFrame {

private final JLabel imageView;

private Mat srcMat;

public InPaint() {

srcMat = Imgcodecs.imread("./lena_dirty.jpg");

JLabel inpaintRadiusLabel = new JLabel("半径:" + 0);

inpaintRadiusLabel.setBounds(15, 10, 60, 15);

JSlider inpaintRadiusBar = new JSlider(0, 50);

inpaintRadiusBar.setValue(0);

inpaintRadiusBar.setBounds(80, 5, 120, 25);

JRadioButton nsBtn = new JRadioButton("INPAINT_NS");

nsBtn.setSelected(true);

nsBtn.setBounds(230, 10, 120, 15);

JRadioButton teleaBtn = new JRadioButton("INPAINT_TELEA");

teleaBtn.setBounds(350, 10, 120, 15);

ButtonGroup group = new ButtonGroup();

group.add(nsBtn);

group.add(teleaBtn);

inpaintRadiusBar.addChangeListener(new ChangeListener() {

@Override

public void stateChanged(ChangeEvent e) {

double value = inpaintRadiusBar.getValue();

inpaintRadiusLabel.setText("半径:" + value);

int type = nsBtn.isSelected() ? 1 : 0;

inpaint(value, type);

}

});

nsBtn.addItemListener(new ItemListener() {

@Override

public void itemStateChanged(ItemEvent e) {

double value = inpaintRadiusBar.getValue();

inpaintRadiusLabel.setText("半径:" + value);

int type = nsBtn.isSelected() ? 1 : 0;

inpaint(value, type);

}

});

teleaBtn.addItemListener(new ItemListener() {

@Override

public void itemStateChanged(ItemEvent e) {

double value = inpaintRadiusBar.getValue();

inpaintRadiusLabel.setText("半径:" + value);

int type = nsBtn.isSelected() ? 1 : 0;

inpaint(value, type);

}

});

imageView = new JLabel();

imageView.setBounds(5, 5, 512, 512);

this.setTitle("图片处理");

this.setSize(550, 580);

this.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);

this.getContentPane().setLayout(null);

this.getContentPane().add(inpaintRadiusLabel);

this.getContentPane().add(inpaintRadiusBar);

this.getContentPane().add(nsBtn);

this.getContentPane().add(teleaBtn);

this.getContentPane().add(imageView);

setIcon(srcMat);

}

/**

* 区域填充

* @param inpaintRadius 数值越大，修复区域越模糊-过渡越好

* @param flags 本例 INPAINT_TELEA 慢于 INPAINT_NS

private void inpaint(double inpaintRadius, int flags) {

Mat dstMat = new Mat(srcMat.size(), CvType.CV_8UC3);

Mat inpaintMaskMat = Imgcodecs.imread("./lena_dirtymask.jpg", CvType.CV_8UC1); // 模拟手动选择的要修复的区域

if (1 == flags) {

Photo.inpaint(srcMat, inpaintMaskMat, dstMat, inpaintRadius, Photo.INPAINT_NS); // 边缘线算法

} else {

Photo.inpaint(srcMat, inpaintMaskMat, dstMat, inpaintRadius, Photo.INPAINT_TELEA); // 临近像素算法

}

setIcon(dstMat);

}

// 去色

private void decolor() {

srcMat = Imgcodecs.imread("./lena.jpg");

Mat grayscaleMat = new Mat(); // 灰阶图，8位单通道

Mat colorBoostMat = new Mat(); // 彩图，8位三通道

Photo.decolor(srcMat, grayscaleMat, colorBoostMat);

setIcon(grayscaleMat);

setIcon(colorBoostMat);

}

// 降噪-去除杂点。图片整体上有种柔化的感觉

private void fastNlMeansDenoising(){

Mat dstMat = new Mat();

// 方法1。可用于单通道、3通道图

// fastNlMeansDenoising(Mat src, Mat& dst, float h = 3, int templateWindowSize = 7, int searchWindowSize = 21)

Photo.fastNlMeansDenoising(srcMat, dstMat);

// 方法2。按照索引去除杂点。可用于单通道、3通道图

Mat im1 = Imgcodecs.imread("./lena1.jpg",Imgcodecs.CV_LOAD_IMAGE_COLOR);

Mat im2 = Imgcodecs.imread("./lena2.jpg",Imgcodecs.CV_LOAD_IMAGE_COLOR);

List<Mat> srcImgs=new ArrayList<>();

srcImgs.add(im1); // 杂点图片

srcImgs.add(im2); // 杂点图片

srcImgs.add(srcMat);

// fastNlMeansDenoisingMulti(vector_Mat srcImgs, Mat& dst, int imgToDenoiseIndex, int temporalWindowSize, float h = 3, int templateWindowSize = 7, int searchWindowSize = 21)

// imgToDenoiseIndex 指srcImgs中"杂点的索引"图片的索引。

// temporalWindowSize 要去除的杂点周围的像素数量，须奇数。

Photo.fastNlMeansDenoisingMulti(srcImgs, dstMat, 2, 1);

// 方法3。可用于3通道图

// fastNlMeansDenoisingColored(Mat src, Mat& dst, float h = 3, float hColor = 3, int templateWindowSize = 7, int searchWindowSize = 21)

Photo.fastNlMeansDenoisingColored(srcMat, dstMat);

// 方法4

// fastNlMeansDenoisingColoredMulti(vector_Mat srcImgs, Mat& dst, int imgToDenoiseIndex, int temporalWindowSize, float h = 3, float hColor = 3, int templateWindowSize = 7, int searchWindowSize = 21)

Photo.fastNlMeansDenoisingColoredMulti(srcImgs, dstMat, 2, 1);

// 方法5

Mat img = Imgcodecs.imread("./lena.jpg",Imgcodecs.CV_LOAD_IMAGE_GRAYSCALE); // 转灰度图 CV_LOAD_IMAGE_GRAYSCALE ：gray = 0.299 * r + 0.587 * g + 0.114 * b ，小数点部分四舍五入为整数

Mat img1 = Imgcodecs.imread("./lena1.jpg",Imgcodecs.CV_LOAD_IMAGE_GRAYSCALE);

Mat img2 = Imgcodecs.imread("./lena2.jpg",Imgcodecs.CV_LOAD_IMAGE_GRAYSCALE);

List<Mat> srcList=new ArrayList<Mat>();

srcList.add(img);

// srcList.add(img1);

// srcList.add(img2);

// denoise_TVL1(vector_Mat observations, Mat result, double lambda = 1.0, int niters = 30)

// lambda 数值越小，图像越模糊-柔化

// niters 运算次数

Photo.denoise_TVL1(srcList, dstMat, 10, 10);

setIcon(dstMat);

}

private void setIcon(Mat mat) {

BufferedImage image = CVUtil.matToBufferedImage(mat); // CVUtil 见 http://www.gaohaiyan.com/3229.html

imageView.setIcon(new ImageIcon(image));

}

- end

声明

本文由 cuiweiyou 原创，转载请注明出处：http://www.gaohaiyan.com/3551.html

承接App定制、企业web站点、办公系统软件设计开发，外包项目，毕设

opencv

OpenCV和Java，图片形变：缩放
OpenCV和Java，图片滤镜、平滑、浮雕、边缘检测
OpenCV和Java，点、线、矩、形、圆、椭圆、不规则多边形、中英文字，填色
OpenCV和Java，光学变换：腐蚀、膨胀、形态学变换、边缘检测