-
Book Overview & Buying
-
Table Of Contents
Hands-On Image Processing and Computer Vision with Python - Second Edition
By :
Classical segmentation methods rely on explicit mathematical modeling of intensity distributions, spatial continuity, edge information, region homogeneity and energy minimization principles. These approaches are interpretable, computationally efficient, and form the theoretical backbone of modern deep learning models. We group the topics into conceptual blocks.
Thresholding is one of the most fundamental and interpretable approaches in image segmentation. It partitions the intensity space into disjoint regions such that pixels within each region share similar gray-level characteristics. While classical thresholding methods separate an image into two classes (foreground/background), many real-world images contain multiple semantic regions, motivating multi-level thresholding.
The classical Otsu’s method (proposed by Nobuyuki Otsu in 1979) determines an optimal...
Change the font size
Change margin width
Change background colour