Opencv learning notes -day 12 (ROI region extraction and inrange() function operation)

day12-roi Region extraction

utilize mask( Mask ) Technology to extract solid color background image ROI People and things in the area , And add the extracted person or object to other images .
Through the first cvtColor() function , The original RGB Color image conversion to hsv Images in color space （ High color contrast ）, And then through inRange() Function to obtain ROI Regional Mask, recycling bitwise_not() Function negates and extracts ROI Area .
Finally, the original image and ROI Region images are overlapped to achieve the purpose of extraction .

function

1. inRange()

inRange() Function can realize the binarization function , Multiple channels can be operated at the same time
The main function is to set the pixel value within two thresholds to white （255）, The pixel value not within the threshold interval is set to black （0）

// Definition 
CV_EXPORTS_W void inRange(InputArray src, InputArray lowerb,
                          InputArray upperb, OutputArray dst);

inRange( Input , Upper boundary , Lower boundary , Output )

Parameters	explain
src	InputArray
lowerb	The upper boundary threshold to be extracted
upperb	The threshold value of the lower boundary to be extracted
dst	OutputArray

example 1

inRange(hsv,Scalar(35,43,46),Scalar(77,255,255),mask);
// Deduct the green part of the picture 

In the above example Scalar(35,43,46) yes green The lower boundary of ,Scalar(77,255,255) yes green The lower boundary of , This statement sets the pixels within these two thresholds to white （255）, Pixels outside the threshold are set to black （0）
The display effect is shown in the figure

Supplement the boundary threshold of other colors ：

Code

quickopencv.h

#pragma once

#include <opencv2\highgui.hpp>

#include <opencv2\imgproc.hpp>

using namespace cv;
// Defining classes 
class QuickDemo{
    
public:
	void colorSpace_Demo(Mat &image);// Color space conversion function 2021-12-24
	void mat_creation_demo(Mat &image);//Mat Object and creation 2021-12-27
	void pixel_vist_Demo(Mat &image);// Read and write operation of image pixels 2022-1-3
	void operators_demo(Mat &image);// Arithmetic operation of image pixels 2022-1-4
	void tracking_bar_demo(Mat &image);// Scroll bar operation demonstration 2022-1-7
	void key_demo(Mat &image);// Keyboard response operation 2022-1-12
	void color_style_demo(Mat &image);//OpenCV With color table operation 2022-1-12
	void bitwise_demo(Mat &image);// Bit operation （ Logical operation ）2022-1-17
	void channels_demo(Mat &image);// Channel separation and merge 2022-1-17
	void inrange_demo(Mat &image);//roi Region extraction  2022-1-20
};

QuickDemo.cpp

#include <opencv2\highgui.hpp>

#include <opencv2\imgproc.hpp>

#include<quickopencv.h>

#include <iostream>
void QuickDemo::inrange_demo(Mat &image)
{
    
	Mat hsv;
	cvtColor(image, hsv, COLOR_BGR2HSV);// First of all, will bgr convert to hsv
	Mat mask;
	inRange(hsv,Scalar(35,43,46),Scalar(77,255,255),mask);// Deduct the green part of the picture 
	//inRange( Input , Upper boundary , Lower boundary , Output )
	// Check whether the array elements are in the other two arrays ( Upper boundary , Lower boundary ) Between element values , Upper boundary Scalar(35,43,46) Is the green upper boundary ,
	// Lower boundary Scalar(77,255,255) Is the green lower boundary , The function sets the pixel value within two boundary thresholds to white （255）,
	// The pixel value not within the threshold interval is set to black （0）  That is, binarization 
	imshow("mask", mask);// Show results 1  Green screen background setting 255  To extract 0
	bitwise_not(mask, mask);// Invert inverts the binary image , About to extract some settings 255
	imshow("mask_2", mask);// Show results 2  After reversing 
	Mat back = Mat::zeros(image.size(), image.type());// Black background 
	image.copyTo(back, mask);// hold mask and image After overlapping, put mask The pixel value in is 0（black） The point corresponding to image The point in the becomes transparent , And keep other points 
	// The effect is to keep the part to be extracted 
	imshow("roi Region extraction ", back);// Show results 3 roi Extract the results 
}

#include <iostream>
#include <opencv2\highgui.hpp>
#include <opencv2\imgproc.hpp>
#include<quickopencv.h>

using namespace cv;
using namespace std;


int main()
{
    
	Mat scr = imread("...\\pictures\\image\\ Green screen Pikachu .jfif");// Open a picture 
	if (!scr.data == 1)// Sentenced to empty 
		return -1;
	namedWindow(" window 1", WINDOW_NORMAL);// establish  WINDOW_FREERATIO window 
	imshow(" window 1",scr);// Show in the created window 
	QuickDemo qd;
	qd.inrange_demo(scr);
	waitKey(0);
	return 0;
}

Original picture

Show results 1 Green screen background setting 255 To extract 0

Show results 2 After reversing The part to be extracted is set to 255

Show results 3 roi Extract the results