Image Segmentation

Image Segmentation is the process of dividing an image into segments or unique areas of interest in an image. Segmentation is done in two main ways:

1. Connecting a series of detected edges
2. Grouping an image into separate regions by area or distinct trait

1. Image Contours

Image contours are continuous curves that follow the edges along a perceived boundary. Contours also provide a lot of information about the shape of an object boundary.

Edge detection algorithms are often used to detect the boundaries of objects. But, after performing edge detection you'll often be left with sets of edges that highlight not only object boundaries, but also interesting features and line. And to do image segmentation, you want only boundaries that mark distinct areas and objects in an image. See the example below:

Image contouring is great for detecting documents:

2. Hough Transform

The Hough transform is a technique which can be used to isolate features of a particular shape within an image.

In the this exercise, we will see how hough transform is used to detect objects (like a phone) and car lanes.

3. CNNs

• Convolutional Neural Networks: One of the most basic use of CNN's in image segmentation is actually an extended classification algorithm. Instead of recognizing single objects; CNN's can be trained to classify every pixel in an image so that the image is broken up into these different segments. This can be a very slow learning process, so other, faster algorithms are still being researched and developed. Some networks learn from smaller maps of image features, which we'll learn more about soon. You're encouraged to read about the latest CNN segmentation techniques, here.

4. Other image segmentation algorithms:

• K-Means Clustering: it is an unsupervised learning algorithm that separates an image into segments by clustering data points that have similar traits