Virtual notepad using opencv

Virtual notepad using opencv 

virtual notepad using OpenCV involves detecting hand gestures or finger movements to draw on a virtual canvas. Below is a simple example that tracks a specific color (e.g., the tip of a finger) to draw on a canvas. You can use this basic approach as a foundation for more complex implementations.

Step 1: Install Required Libraries

pip install opencv-python numpy

Step 2: Write the Virtual Notepad Code

import cv2
import numpy as np

# Function to detect a specific color (e.g., the tip of a colored marker or finger) in the frame
def detect_color(frame, lower_color, upper_color):
    hsv_frame = cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)
    mask = cv2.inRange(hsv_frame, lower_color, upper_color)
    return mask

# Initialize the webcam
cap = cv2.VideoCapture(0)

# Define the range of the color you want to track (e.g., blue)
lower_blue = np.array([100, 150, 0])
upper_blue = np.array([140, 255, 255])

# Create a blank image for the notepad
canvas = np.zeros((480, 640, 3), dtype=np.uint8)

# Previous x, y position of the point
prev_x, prev_y = None, None

while True:
    ret, frame = cap.read()
    if not ret:
        break
    
    # Flip the frame to avoid mirror effect
    frame = cv2.flip(frame, 1)
    
    # Detect the color in the frame
    mask = detect_color(frame, lower_blue, upper_blue)
    
    # Find contours in the mask
    contours, _ = cv2.findContours(mask, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
    
    # If any contour is found, consider it as the point to draw
    if contours:
        # Find the largest contour by area
        largest_contour = max(contours, key=cv2.contourArea)
        ((x, y), radius) = cv2.minEnclosingCircle(largest_contour)
        
        # Draw if the detected contour is large enough
        if radius > 10:
            if prev_x is not None and prev_y is not None:
                # Draw a line on the canvas
                cv2.line(canvas, (prev_x, prev_y), (int(x), int(y)), (255, 0, 0), 5)
            
            # Update previous positions
            prev_x, prev_y = int(x), int(y)
    
    # Combine the frame and canvas
    combined = cv2.addWeighted(frame, 0.5, canvas, 0.5, 0)
    
    # Display the result
    cv2.imshow("Virtual Notepad", combined)
    
    # Clear the canvas when 'c' is pressed
    if cv2.waitKey(1) & 0xFF == ord('c'):
        canvas = np.zeros((480, 640, 3), dtype=np.uint8)
        prev_x, prev_y = None, None
    
    # Break the loop when 'q' is pressed
    if cv2.waitKey(1) & 0xFF == ord('q'):
        break

# Release the resources
cap.release()
cv2.destroyAllWindows()

How the Code Works:

1. Color Detection: The detect_color function converts the video frame to HSV color space and applies a mask to detect a specific color. In this example, we’re tracking a blue color.
2. Contours: The code finds contours in the masked image and selects the largest contour to represent the drawing point.
3. Drawing: As the detected point moves, the previous and current positions are connected with a line on the canvas.
4. Canvas: A blank image (canvas) is used as the drawing surface, which is combined with the live video feed to display the drawing.
5. Reset and Exit: Press 'c' to clear the canvas or 'q' to exit the application.

Step 3: Run the Code

When you run the code, a window will open showing the live feed from your webcam. You can use an object (like a blue pen) to draw on the virtual canvas. Press 'c' to clear the canvas and 'q' to quit the application.

Notes:

• You can adjust the lower_blue and upper_blue values to track a different color or fine-tune the current color detection.
• This example uses a simple color-based approach for drawing. For more advanced gesture recognition, you can integrate hand-tracking models.