python - How can I detect and track people using OpenCV? -


I have a camera that will be stationary, indicating that inside the home area people will run within the last 5 meters of the camera. Using the OpenCV , I want to find people walking in the past - the array of people finding my ideal return, along with the rectangles.

I have many built-in samples:

  • Any Python samples are not actually implemented
  • C < The robot tracking sample looks promising, but it does not accept live video, which makes difficult tests, it is also the most complex of samples, which helps in removing related knowledge and making it a python API problematic.
  • The C sample also looks promising, it calculates a silhouette with subsequent video frames, possibly I find that strongly connected components and individual blobs and their I can use it to remove the box - but I am still trying to find a way to identify the blobes found in the later frames as a similar blob.

Is anyone capable of providing guidance or samples to do this - primarily Python ? OpenCV's latest SVN version includes (compulsory) implementation of a HOG-based pedestrian identity.

It comes with a pre-trained detector and python cover. The original usage is as follows:

  to cv import * storage = CreateMemStorage (0) img = LoadImage (file) # or list from camera = HOGDetectMultiScale (IMG, storage, win_stride = (8 , 8), padding = (32,32), scale = 1.05, group_treadhold = 2))  

So instead of tracking, you can run the detector only in each frame and using Its output directly

See samples / python / peopledetect.py for more complete dragon examples (both opencv sources) for implementation src / cvaux / cvhog.cpp In).


Comments

Popular posts from this blog

sql - dynamically varied number of conditions in the 'where' statement using LINQ -

asp.net mvc - Dynamically Generated Ajax.BeginForm -

Debug on symbian -