Home > Software engineering >  Do you want to be continuous frame pedestrian tracking what idea to have excuse me?
Do you want to be continuous frame pedestrian tracking what idea to have excuse me?

Time:09-21

Now identify each image on the pedestrian but how am I to judge the frame of this and the next frame everyone who this person is the same person?
Must be able to identify until the match point this way because the picture is too small give up after just tried
Pray god give some ideas

CodePudding user response:

To determine whether there are pedestrians should be relatively simple, the two image subtraction, if determine whether the same pedestrians is not easy

CodePudding user response:

What color do what characteristics calculation, what

CodePudding user response:

Do you want to do is face recognition + motion detection?

CodePudding user response:

Fyi:
 # "Object Tracking using OpenCV (c + +/Python)" 
The import cv2
The import sys

(major_ver minor_ver, subminor_ver)=(cv2. __version__). The split ('. ')

If __name__=="__main__ ':

# Set up the tracker.
Home # MIL, you can also use the

Tracker_types=[' BOOSTING ', 'MIL', 'KCF', 'TLD', 'MEDIANFLOW', 'GOTURN', 'MOSSE]
Tracker_type=tracker_types [2]

If the int (minor_ver) & lt; 3:
The tracker=cv2. Tracker_create (tracker_type)
The else:
If tracker_type=='BOOSTING:
The tracker=cv2. TrackerBoosting_create ()
If tracker_type=='MIL:
The tracker=cv2. TrackerMIL_create ()
If tracker_type=='KCF:
The tracker=cv2. TrackerKCF_create ()
If tracker_type=='TLD:
The tracker=cv2. TrackerTLD_create ()
If tracker_type=='MEDIANFLOW:
The tracker=cv2. TrackerMedianFlow_create ()
If tracker_type=='GOTURN:
The tracker=cv2. TrackerGOTURN_create ()
If tracker_type=='MOSSE:
The tracker=cv2. TrackerMOSSE_create ()

# Read video
Video=cv2. VideoCapture (" stage. Mp4 ")

# Exit if video not the opened.
If not video. IsOpened () :
Print (" Could not open the video ")
Sys. The exit ()

For the f in the range 25 * (10) :
Ok, frame=video. The read ()
If not ok:
Print (' always read a video file ')
Sys. The exit ()


# Uncomment the line below to select a company's bounding box
Bbox=cv2. SelectROI (frame, False)
Cv2. DestroyWindow (' ROI selector ')
Print (" selectROI: "+ STR (bbox))
If bbox==(0,0,0,0) :
# Define an default bounding box
Bbox=(700, 255, 192, 473)
Print (" You canceled selection. Use the default "+ STR (bbox))

# the Initialize the tracker with the first frame and bounding box
Ok=tracker. The init (frame, bbox)
Print (ok)

While True:
# Read a new frame
Ok, frame=video. The read ()
If not ok:
Break

# Start timer
The timer=cv2. GetTickCount ()

# the Update tracker
Ok, bbox=tracker. Update (frame)

# Calculate Frames per second (FPS)
FPS=cv2. GetTickFrequency ()/(cv2. The getTickCount () - timer);

# the Draw bounding box
If ok:
# Tracking success
P1=(int (bbox [0]), int (bbox [1]))
P2=(int (bbox bbox [0] + [2]), int (bbox bbox [1] + [3]))
Cv2. A rectangle (frame, p1, p2, (0, 255), 2, 1)
The else:
# Tracking failure
Cv2. PutText (frame, "Tracking failure detected", (100), cv2. FONT_HERSHEY_SIMPLEX, 0.75, (0,0,255), 2)

# Display the tracker type on frame
Cv2. PutText (frame, tracker_type + "Tracker," (100), cv2. FONT_HERSHEY_SIMPLEX, 0.75, (50170, 50), 2);

# Display FPS on frame
Cv2. PutText (frame, "FPS:" + STR (int (FPS)), (100, 50), cv2. FONT_HERSHEY_SIMPLEX, 0.75, (50170, 50), 2);

# the Display result
Cv2. Imshow (" Tracking ", frame)

# Exit if ESC pressed
K=cv2. WaitKey (1) & amp; 0 XFF
If k==27:
Break
If k==32:
Bbox1=cv2. SelectROI (frame, False)
Cv2. DestroyWindow (' ROI selector ')
Print (" selectROI: "+ STR (bbox1))
If bbox1==(0,0,0,0) :
Print (" You canceled selection. Use the old "+ STR (bbox))
The else:
Bbox=bbox1
Print (" new selectROI: "+ STR (bbox))
Del tracker
The tracker=cv2. TrackerKCF_create ()
Ok=tracker. The init (frame, bbox)
Print (ok)
Cv2. DestroyWindow (' Tracking ')

CodePudding user response:

Use opencv is very simple, just above, use C calls, hundreds of lines of code

CodePudding user response:

"OpenCV computer visual programming strategy (3rd edition)" in chapter 13 tracking moving targets

CodePudding user response:

quoted zhao 4, 4/f, the teacher's reply:
are for reference only:
 # "Object Tracking using OpenCV (c + +/Python)" 
The import cv2
The import sys

(major_ver minor_ver, subminor_ver)=(cv2. __version__). The split ('. ')

If __name__=="__main__ ':

# Set up the tracker.
Home # MIL, you can also use the

Tracker_types=[' BOOSTING ', 'MIL', 'KCF', 'TLD', 'MEDIANFLOW', 'GOTURN', 'MOSSE]
Tracker_type=tracker_types [2]

If the int (minor_ver) & lt; 3:
The tracker=cv2. Tracker_create (tracker_type)
The else:
If tracker_type=='BOOSTING:
The tracker=cv2. TrackerBoosting_create ()
If tracker_type=='MIL:
The tracker=cv2. TrackerMIL_create ()
If tracker_type=='KCF:
The tracker=cv2. TrackerKCF_create ()
If tracker_type=='TLD:
The tracker=cv2. TrackerTLD_create ()
If tracker_type=='MEDIANFLOW:
The tracker=cv2. TrackerMedianFlow_create ()
nullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnullnull
  • Related