Implementation-of-Filters

Aim:

To implement filters for smoothing and sharpening the images in the spatial domain.

Software Required:

Anaconda - Python 3.7

Algorithm:

Step 1:

Import the required libraries and read the image.

Step 2:

Convert the saved BGR image to RGB using cvtColor().

Step 3:

Use the filters required for image smoothing and sharpening.

Step 4:

Apply the filters using cv2.filter2D() for each respective filters.

Step 5:

Plot the images of the original one and the filtered one using plt.figure() and cv2.imshow().

Program:

Developed By   : Gunaseelan G
Register Number: 212221230031

import cv2
import numpy as np
import matplotlib.pyplot as plt
image=cv2.imread('bike.png')
image2=cv2.cvtColor (image1, cv2.COLOR_BGR2RGB)

1. Smoothing Filters :

i) Using Averaging Filter:

kernel1 = np.ones((11,11),np.float32)/121
avg_filter = cv2.filter2D(image2,-1,kernel1)
plt.figure(figsize = (9,9))
plt.subplot(1,2,1)
plt.imshow(image2)
plt.title("Original")
plt.axis("off")
plt.subplot(1,2,2)
plt.imshow(avg_filter)
plt.title("Filtered")
plt.axis("off")

ii) Using Weighted Averaging Filter:

kernel2 = np.array([[1,2,1],[2,4,2],[1,2,1]])/16
weighted_filter = cv2.filter2D(image2,-1,kernel2)
plt.figure(figsize = (9,9))
plt.subplot(1,2,1)
plt.imshow(image2)
plt.title("Original")
plt.axis("off")
plt.subplot(1,2,2)
plt.imshow(weighted_filter)
plt.title("Filtered")
plt.axis("off")

iii) Using Gaussian Filter:

gaussian_blur = cv2.GaussianBlur(src = image2, ksize = (11,11), sigmaX=0, sigmaY=0)
plt.figure(figsize = (9,9))
plt.subplot(1,2,1)
plt.imshow(image2)
plt.title("Original")
plt.axis("off")
plt.subplot(1,2,2)
plt.imshow(gaussian_blur)
plt.title("Filtered")
plt.axis("off")

iv) Using Median Filter:

median = cv2.medianBlur(src=image2,ksize = 11)
plt.figure(figsize = (9,9))
plt.subplot(1,2,1)
plt.imshow(image2)
plt.title("Original")
plt.axis("off")
plt.subplot(1,2,2)
plt.imshow(median)
plt.title("Filtered (Median)")
plt.axis("off")

2. Sharpening Filters:

i) Using Laplacian Kernal:

kernel3 = np.array([[0,1,0],[1,-4,1],[0,1,0]])
laplacian_kernel = cv2.filter2D(image2,-1,kernel3)
plt.figure(figsize = (9,9))
plt.subplot(1,2,1)
plt.imshow(image2)
plt.title("Original")
plt.axis("off")
plt.subplot(1,2,2)
plt.imshow(laplacian_kernel)
plt.title("Filtered (Laplacian Kernel)")
plt.axis("off")

ii) Using Laplacian Operator:

laplacian_operator = cv2.Laplacian(image2,cv2.CV_64F)
plt.figure(figsize = (9,9))
plt.subplot(1,2,1)
plt.imshow(image2)
plt.title("Original")
plt.axis("off")
plt.subplot(1,2,2)
plt.imshow(laplacian_operator)
plt.title("Filtered (Laplacian Operator)")
plt.axis("off")

Output:

Smoothing Filters:

i) Using Averaging Filter:

ii) Using Weighted Average Filter:

iii) Using Gaussian Filter:

iv) Using Median Filter:

Sharpening Filters:

i) Using Laplacian Kernel:

ii) Using Laplacian Operator:

Result:

Thus the filters are designed for smoothing and sharpening the images in the spatial domain.

guru-guna / implementation-of-filters Goto Github PK