Opencv implements object tracking

1. Contour finding and contour drawing

https://mydreamambitious.blog.csdn.net/article/details/125339604

---

2. Gaussian denoising

https://mydreamambitious.blog.csdn.net/article/details/125203273

---

3. Corrosion and expansion

https://mydreamambitious.blog.csdn.net/article/details/125265431

---

4.cv2.inRange function

effect ：rgb Color space to hsv Color space , The color removes the white background
Parameters ：
The first parameter ：hsv Refers to the input original drawing ;
The second parameter ：lower_red It's below this in the image lower_red Value , The pixel value becomes 0;
The third parameter ：upper_red It's higher than this in the image upper_red Value , The pixel value becomes 0;

---

5.cv2.minEnclosingCircle function

Function function ： Get the smallest circle containing two-dimensional point set
Parameters ：
InputArray points： Input two-dimensional point set ;
Point2f& center： Represents the center coordinate of the output circle , yes float type ;
float& radius： The radius of the smallest circle output , yes float type ;

---

6.cv2.moments function

Find the moment of the contour ：
mu = cv.moments(contour) #contour It's an outline
cx = mu[‘m10’] / mu[‘m00’] # outline contour Abscissa of the center of mass of
cy = mu[‘m01’] / mu[‘m00’] # outline contour The ordinate of the center of mass of
https://blog.csdn.net/Caesar6666/article/details/103257632

---

7. Code combat

import os
import cv2
import numpy as np

cap=cv2.VideoCapture(0)
# Set up RGB->HSV Low threshold and high threshold 
Obj_low=np.array([0,0,0])
Obj_high=np.array([179,157,79])

while cap.isOpened():
    OK,frame=cap.read()
    if OK==False:
        break
    # Flip the picture ,Opencv The image read is the opposite of our normal display 
    img=cv2.flip(src=frame,flipCode=2)
    img=cv2.resize(src=img,dsize=(600,600))
    # Gaussian processing of the image 
    gauss=cv2.GaussianBlur(src=img,ksize=(3,3),sigmaX=0)
    # Convert the image to grayscale 
    gray=cv2.cvtColor(src=gauss,code=cv2.COLOR_BGR2HSV)
    inrange=cv2.inRange(src=gray,lowerb=Obj_low,upperb=Obj_high)
    # Corrode the image 
    kernel=cv2.getStructuringElement(shape=cv2.MORPH_RECT,ksize=(3,3))
    erode=cv2.erode(src=inrange,kernel=kernel,iterations=2)
    # Inflate the graph 
    dilate=cv2.dilate(src=erode,kernel=kernel,iterations=2)
    # Find the outline 
    contours,hierarchy=cv2.findContours(image=dilate,mode=cv2.RETR_TREE,
                                        method=cv2.CHAIN_APPROX_SIMPLE)
    maxArea=0
    index=0
    length=len(contours)
    if length>0:
        for i in range(length):
            # Calculate the area of the contour 
            area=cv2.contourArea(contours[i])
            if area>maxArea:
                maxArea=area
                index=i
        # Get the circle containing the minimum point set 
        ((x,y),radius)=cv2.minEnclosingCircle(contours[index])
        M=cv2.moments(contours[index])
        center=(int(M['m10']/M['m00']),int(M['m01']/M['m00']))

        if radius>5:
            cv2.circle(img=img,center=center,radius=2,
                       color=(0,255,0),thickness=-1)
            cv2.circle(img=img,center=center,radius=int(radius),
                       color=(0,0,255),thickness=3)

    cv2.imshow('image',img)
    if cv2.waitKey(1)&0xFF==27:
        break
cv2.destroyAllWindows()


if __name__ == '__main__':
    print('Pycharm')