imread function + cvtColor () function

The image is loaded (with cv :: imread)

• imread function is to load an image file into a Mat object that the first parameter represents the image file name
• The second parameter indicates what type of loaded images, support for common values ​​of three parameters

---

1. IMREAD_UNCHANGED (<0) represents the original load, without any change
2. IMREAD_GRAYSCALE (0) represents the original gray-scale image as come loading
3. IMREAD_COLOR (> 0) represents the original RGB image is loaded as come

Note: OpenCV support JPG, PNG, TIFF and other common image file formats loaded

Display images (cv :: namedWindos and cv :: imshow)

• namedWindos OpenCV function is to create a window, it is automatically created by the release of OpenCV with, you do not need to take to destroy it.
• Common usage namedWindow ( "Window Title", WINDOW_AUTOSIZE)
• WINDOW_AUTOSIZE automatically the image size, display size according to the window, the window size can not be artificially changed
• WINDOW_NORMAL, when integrated with the QT will use, it allows you to modify the window size.
• imshow images up to a specified window according to the window name is displayed, the first parameter is the name of the window, the second parameter is an object Mat

Save the image (cv :: imwrite)

• Save the image file to the specified directory path
• Only 8-bit, 16-bit PNG, JPG, Tiff file format and a single-channel or three-channel image BGR can only be saved in this way
• PNG format saved time can be saved alpha channel pictures
• You can specify compression parameters

---

1. The first argument: will save another name for the image, you can define your own name.
2. The second argument: you want to save the image name, is already existing image.
3. Third parameter: the parameter set for the image format generally omitted, do not write

opencv provides cvtColor () function to implement these functions. First look at cvtColor function definition:

C ++: void cvtColor (InputArray the src, outputArray DST, int code, int dstCn = 0);
. 1
of parameters:
InputArray the src: an input image that is to be the original image color space conversion may be Mat class.
OutputArray DST: an output image. i.e. after the color space conversion store images may Mat type
int code:. transcoded or identification, i.e. where it is determined to convert what format images into what format images detail later will
int dstCn = 0:. target image number of channels, if the value is 0, and by the src code decision

code show as below:

//首先是导入相关函数（类似于包java import....）
#include<opencv2\opencv.hpp>
#include<iostream>
#include<math.h>

//域名控件（类似java导入包后，无需在每个对象中添加包名例如：本来要写cv::Mat 添加域名空间后，只需写Mat）
using namespace cv;

//主函数
int main(int argc, char** argv)
{
    //创建一个Mat对象并在里面写入图片数据(图片即存储像素点的矩阵数组)
    Mat srcImage = imread("E:\\VS2015Opencv\\vs2015\\project\\picture\\01.jpg");
    //判断Mat是否有数据。无，打印消息，退出，有就继续（C和java都差不多）
    if (srcImage.empty()) {
        printf("couldn't load image");
        //暂停
        getchar();
        //返回-1,退出
        return -1;
    }
    //显示原图像
    namedWindow("原图像", WINDOW_AUTOSIZE);
    imshow("原图像", srcImage);

    //将图像转换为灰度图，采用CV_前缀
    Mat grayImage;
    cvtColor(srcImage, grayImage, CV_BGR2GRAY);     //将图像转换为灰度图
    namedWindow("灰度图", WINDOW_AUTOSIZE);
    imshow("灰度图", grayImage);

    //将图像转换为HSV，采用COLOR_前缀
    Mat HSVImage;
    cvtColor(srcImage, HSVImage, COLOR_BGR2HSV);    //将图像转换为HSV图
    namedWindow("HSV", WINDOW_AUTOSIZE);
    imshow("HSV", HSVImage);

    waitKey(0);

    return 0;
}