Gesture_Pattern_detection_Model

Simple Python code to track the gesture of movement in front of camera and display the movement pattern.

To run this code you must be sure that you have open cv installed on your system.
I would recommend installing opencv version 4.2.0 because the MOSSE tracker tends to not function in other versions. To install the specefic version of opencv in cmd type pip install opencv-contrib-python==4.2.0.

NOTE:: This code uses open cv version 4.2.0 .

So make sure you have it installed. To install python and pip from ground level go to end of this file.

This code uses cv2 and its MOSSE tracker and matplotlib to plot the gesture pattern.

cap = cv2.VideoCapture(0)

Using front camera to capture gestures.

tracker = cv2.TrackerMOSSE_create()

Creating a MOSSE tracker.

def drawbox(img, bbox):
   .......
   x_coord.append(x)
	 y_coord.append(y)

Defining the function drawbox which is used to draw the box to identify the object to be tracked. Inside the function we update two arrays which store the values of x and y coordinates.

def graph_plot():

In this function we use matplotlib's plot() function to plot the x and y cordinates.

if cv2.waitKey(1) & 0xff == ord('q'):

The tracking stops as we press 'q' button on our keyboard.

To freshly install python and pip::

First check if you have python installed in your computer. Open command prompt and type the command python --version.

If version does not appears that means your system does not have python installed properly. Head over to python.org and download and install python first. While installing be sure to check the Add python to PATH button.

To check the successful installation of python, in cmd type python --version. If the version appears, then it is installed. Newer versions of python have pip preinstalled by defult.

If python is installed then type pip, if it executes properly then pip is present in your system. You should be good to go now.