Udemy Image Processing in Python

Udemy Image Processing in Python

Learn to process, transform and manipulate images at your will by: Smit Shah. Udemy Image Processing in Python. Images are everywhere! We live in a time where images contain lots of information, which is sometimes difficult to obtain. This is why image pre-processing has become a highly valuable skill, applicable in many use cases. In this course, you will learn to process, transform, and manipulate images at your will, even when they come in thousands. You will also learn to restore damaged images, perform noise reduction, smart-resize images, count the number of dots on a dice, apply facial detection, and much more, using scikit-image. After completing this course, you will be able to apply your knowledge to different domains such as machine learning and artificial intelligence, machine and robotic vision, space and medical image analysis, retailing, and many more. Take the step and dive into the wonderful world that is computer vision!

What you'll learn?

In this course, you will learn to process, transform, and manipulate images at your will, even when they come in thousands. You will also learn to restore damaged images, perform noise reduction, smart-resize images, count the number of dots on a dice, apply facial detection, and much more, using scikit-image.

Who this course is for?

Beginner Python developers curious about data science.

Requirements

Basic python knowledge.

Course content

- Make images come alive with scikit-image.

- RGB to grayscale: Import the modules from skimage, Load the rocket image ,Convert the image to grayscale, Show the original image, Show the grayscale image.

- NumPy for images.

- Flipping out: Flip the image vertically, Flip the image horizontally, The original, flipped image, Show the resulting image.

- Histograms:  Obtain the red channel, Plot the red histogram with bins in a range of 256, Set title and show.

- Getting started with thresholding.

- Apply global thresholding: Apply global thresholding, Make the image grayscale using rgb2gray, Obtain the optimal threshold value with otsu, Apply thresholding to the image, Show the image.

- When the background isn't that obvious: Import the otsu threshold function, Obtain the optimal otsu global thresh value, Obtain the binary image by applying global thresholding, Show the binary image obtained.

- Trying other methods: Import the try all function, Import the rgb to gray convertor function, Turn the fruits image to grayscale, Use the try all method on the grayscale image, Show the resulting plots.

- Apply thresholding: Import threshold and gray convertor functions, Turn the image grayscale, Obtain the optimal thresh, Obtain the binary image by applying thresholding, Show the resulting binary image.

- Jump into filtering.

- Edge detection: Import the color module, Import the filters module and sobel function, Make the image grayscale, Apply edge detection filter, Show original and resulting image to compare.

- Blurring to reduce noise: Import Gaussian filter, Apply filter, Show original and resulting image to compare.- Contrast enhancement.

- Aerial image: Import the required module, Use histogram equalization to improve the contrast, Show the original and resulting image.

- Let's add some impact and contrast: Import the necessary modules, Load the image, Apply the adaptive equalization on the original image, Compare the original image to the equalized.

- Transformations.

- Enlarging images: Import the module and function to enlarge images, Import the data module, Load the image from data, Enlarge the image so it is 4 times bigger, Show original and resulting image.

- Proportionally resizing: Import the module and function, Set proportional height so its half its size, Resize using the calculated proportional height and width, Show the original and rotated image.

- Morphology.

- Handwritten letters: Import the morphology module, Obtain the eroded shape, See results.

- Improving thresholded image: Import the module, Obtain the dilated image, See results.

- Image restoration.

- Removing logos: Initialize the mask, Set the pixels where the logo is to 1, Apply inpainting to remove the logo.

- Noise.

- Let's make some noise!  Import the module and function, Add noise to the image, Show original and resulting image.

- Reducing noise: Import the module and function, Apply total variation filter denoising, Show the noisy and denoised images.

- Reducing noise while preserving edges: Import bilateral denoising function, Apply bilateral filter denoising, Show original and resulting images.

- Superpixels & segmentation.

- Superpixel segmentation: Import the slic function from segmentation module, Import the label2rgb function from color module, Obtain the segmentation with 400 regions, Put segments on top of original image to compare.

- Finding contours.

- Contouring shapes: Import the modules, Obtain the horse image, Find the contours with a constant level value of 0.8, Shows the image with contours found.

- Count the dots in a dice's image: Create list with the shape of each contour, Set 50 as the maximum size of the dots shape, Count dots in contours excluding bigger than dots size, Shows all contours found, Print the dice's number.

- Finding the edges with Canny.

- Edges: Import the canny edge detector, Convert image to grayscale, Apply canny edge detector, Show resulting image.

- Right around the corner.

- Perspective:  Import the corner detector related functions and module, Convert image from RGB-3 to grayscale, Apply the detector to measure the possible corners, Find the peaks of the corners using the Harris detector, Show original and resulting image with corners detected.

- Face detection.

- Is someone there? Load the trained file from data, Initialize the detector cascade, Detect faces with min and max size of searching window, Show the detected faces.

- Multiple faces: Load the trained file from data, Initialize the detector cascade, Detect faces with scale factor to 1.2 and step ratio to 1, Show the detected faces.

- Segmentation and face detection: Obtain the segmentation with default 100 regions, Obtain segmented image using label2rgb, Detect the faces with multi scale method, Show the detected faces.

- Real-world applications.

- Privacy protection: Detect the faces, For each detected face, Obtain the face rectangle from detected coordinates, Apply gaussian filter to extracted face, Merge this blurry face to our final image and show it.

- Help Sally restore her graduation photo: Import the necessary modules, Transform the image so it's not rotated, Remove noise from the image, using the chambolle method, Reconstruct the image missing parts.

Udemy Image Processing in Python