Object tracking using tracker in opencv

1. Pre knowledge

（1） Object tracking in the previous article

https://mydreamambitious.blog.csdn.net/article/details/125471084

（2）Opencv Draw a straight line in , rectangular , round , The ellipse , polygon （ Include polygon fill ）, Draw text

https://mydreamambitious.blog.csdn.net/article/details/125392536

（3）Opencv Basic knowledge points in

https://mydreamambitious.blog.csdn.net/article/details/125351256

---

2. Code combat

（1） create a window

# create a window 
cv2.namedWindow(winname='track',flags=cv2.WINDOW_AUTOSIZE)
cv2.resizeWindow(winname='track',width=750,height=600)

（2） Instantiate tracker and tracking algorithm

# Create trackers 
trackers=cv2.MultiTracker_create()
# The algorithm used KCF
track_algorithm=cv2.TrackerKCF_create()

（3） Scales the size of the image

# Scale the height and width of the picture 
    height,width,channel=frame.shape
    # Zoom factor 
    factor=500/width
    height=int(factor*height)
    frame=cv2.resize(src=frame,dsize=(500,height))

（4） Add tracked objects

#  If you press t, Draw a rectangle on the object to be tracked 
    if key==ord('t'):
        # selectRoi A tuple is returned [min_x,min_y,w,h]
        #shwoCrosshair: Indicates whether to draw a cross on the rectangle ;fromCenter: Indicates whether to draw from the center of the rectangle 
        box=cv2.selectROI(windowName='track',img=frame,
                          showCrosshair=True,fromCenter=False)
        # Add tracker 
        trackers.add(track_algorithm,frame,box)

（5） Update the tracking results and draw a rectangle

# Update tracking results 
    ret,boxes=trackers.update(frame)
    endtime=time.time()
    FPS=1/(endtime-startime)
    if ret:
        for box in boxes:
            x,y,w,h=int(box[0]),int(box[1]),int(box[2]),int(box[3])
            cv2.rectangle(img=frame,pt1=(x,y),pt2=(x+w,y+h),color=(0,255,0),thickness=2)

    cv2.putText(img=frame, text='FPS: ' + str(int(FPS)), org=(20, height - 50),
            fontFace=cv2.FONT_HERSHEY_SIMPLEX, fontScale=1.0
            , color=(0, 255, 0), thickness=2)
    cv2.imshow('track',frame)
    key=cv2.waitKey(30)

3. Demo video

Target tracking demo