Region growing refers to a class of segmentation algorithms where a pixel's neighbor is considered to be in the same segment if its intensity is similar to the current pixel. How the similarity is defined varies across algorithms. The initial set of pixels are called seed points that are usually manually selected. The next code block demonstrates how to use the SimpleITK library's implementation of the ConnectedThreshold (on variant of the region growing segmentation algorithm). A cranial MRI scan (T1) medical image is used as the input image. In the case of the ConnectedThreshold algorithm, a voxel's (volume pixel) neighbor is considered to be in the same segment if the neighboring voxel's intensity is within ...
Region growing with SimpleITK
Get Hands-On Image Processing with Python now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.