2. ALGORITHM:{DIP SECTION}
• Step 1: Image Acquisition:Acquire image via camera or stored files (this is done using OpenCV to
open webcam or uses files in a pre recorded image set.Since we are trying for a live demonstration,we
would prefer a webcam
• Step 2: Perform Greyscale processing on acquired image:We are performing image greyscaling to
eliminate every form of color from the image to only leave different shades of grey.This allows us to
use a method of image matching that reduces computational complexities and colour related
complexities and calibration
• Step 3: Perform Gaussian Blurring on Greyscaled Image:We use Gaussian blurring techniques to
remove any excess noise present in the image and smoothen the image. Gaussian blur is also useful
for reducing chromatic aberration, those coloured fringes at high-contrast edges in an image.
• Step 4:Thresholding:Find the intensity gradients of the image.Apply gradient magnitude thresholding or
lower bound cut-off suppression to get rid of spurious response to edge detection.
3. Step 5: Perform Canny Edge detection: Using the canny edge detection algorithm to
detect a wide variety of edges in images.
(Noise reduction,gradient calculation,non-maximum suppression,double
threshold,edge tracking by hysteresis)
Step 6: Find the largest Contour :Find largest contour using OpenCV.
Step 7: Match the contour lines of given image with that of the dataset :performing
pixel by pixel and feature matching to obatain an accurate output regarding the
displayed sign
Step 8: Determine the gesture based on Matching accuracy