Creating more filters with OpenCV and Python

#python #tutorial #beginners #datascience

Introduction

Hello! 😀

In this tutorial I will be showcasing some more filters using OpenCV and Python!
This is a continuation of my previous example which can be found here:
https://dev.to/ethand91/creating-various-filters-with-opencvpython-3077

I've already discussed how to create the virtual environment in previous tutorials so I will skip that part.

Well lets get started creating some more filters! 🥳

Vignette Filter

First we will create the vignette filter. The vignette filter is achieved by creating a broad 2D Gaussian kernel.

def vignette(image, level = 2): 
    height, width = image.shape[:2]

    x_resultant_kernel = cv2.getGaussianKernel(width, width/level)
    y_resultant_kernel = cv2.getGaussianKernel(height, height/level)

    kernel = y_resultant_kernel * x_resultant_kernel.T
    mask = kernel / kernel.max()

    image_vignette = np.copy(image)

    for i in range(3):
        image_vignette[:,:,i] = image_vignette[:,:,i] * mask

    return image_vignette

Here we generate the vignette mask using Gaussian kernels, we then generate the result matrix and then apply the mask to each of the image's color channels.

Embossed Filter

The next filter is the embossed filter:

def embossed_edges(image):

    kernel = np.array([[0, -3, -3], [3, 0, -3], [3, 3, 0]])

    image_emboss = cv2.filter2D(image, -1, kernel = kernel)

    return image_emboss

Here we create an array for each of the channels and then apply it to the image via filter2D.

Outline Filter

The next filter is the outline filter:

def outline(image, k = 9): 
    k = max(k, 9)
    kernel = np.array([[-1, -1, -1], [-1, k, -1], [-1, -1, -1]])

    image_outline = cv2.filter2D(image, ddepth = -1, kernel = kernel)

    return image_outline

Similar to the embossed filter but this time we increase the quality of the outlines.

Style Filter

The final filter is one of my personal favorites, the style filter.

def style(image):
    image_blur = cv2.GaussianBlur(image, (5, 5), 0, 0)
    image_style = cv2.stylization(image_blur, sigma_s = 40, sigma_r = 0.1)

    return image_style