Image Segmentation

Image segmentation is the process of partitioning a digital image into multiple segments (sets of pixels or regions), which are similar with respect to some characteristic or computed property, such as color, intensity, or texture [1].

This repository is based on the course Segmentação de Imagens com Python de A à Z offered by the IA Expert Academy.

Intructions:

To read the files in this repository, you can navigate through the directories and files or use the links in this readme file.

Content:

Classical Techniques

1. Thresholding

• 1.1. Global Thresholding

  • 1.1.1. Thresholding Theory

  • 1.1.2. Binary Thresholding

  • 1.1.3. Inverted Binary Thresholding

  • 1.1.4. Truncated Thresholding

  • 1.1.5. To Zero Thresholding

  • 1.1.6. Inverted To Zero Thresholding

  • 1.1.7. Tomography Challenge

  • 1.1.8. Thermal 01 Challenge

  • 1.1.9. Thermal 02 Challenge

  • 1.1.10. Coins Challenge

  • 1.1.11. Otsu's Method Theory

  • 1.1.12. Otsu's Method - Implementation 01

  • 1.1.13. Otsu's Method - Implementation 02

• 1.2. Adaptive Thresholding

  • 1.2.1. Adaptive Thresholding Theory

  • 1.2.2. Adaptive Thresholding - Mean

  • 1.2.3. Adaptive thrsholding - Gaussian

2. Edge Detection

• 2.1. Sobel Operator

• 2.2. Canny Edge

• 2.3. Canny Edge - Coins Challenge

• 2.4. Canny Edge - Thermal Challenge

• 2.5. Canny Edge - Noise Challenge

3. Region-Based Segmentation

• 3.1. Region-Based

• 3.2. Region-Based - Landscape 02 Challenge

• 3.3. Region-Based - Tomography Challenge

• 3.4. Region-Based - Eagle Challenge

4. Cluster-Based Segmentation

• 4.1. Cluster-Based Segmentation - Introduction

• 4.2. Cluster-Based Segmentation - Landscape

• 4.3. Cluster-Based Segmentation - Eagle

• 4.4. Cluster-Based Segmentation - Tomography Challenge

5. Watershed Segmentation

• 5.1. Watershed Segmentation - Introduction

• 5.2. Watershed Segmentation - Coins Challenge

• 5.3. Watershed Segmentation - RBC Challenge

6. Color-Based Segmentation

• 6.1. Color-Based Segmentation - Introduction

• 6.2. Color-Based Segmentation - Magic Cube Challenge RGB

• 6.3. Color-Based Segmentation - Magic Cube Challenge HSV

• 6.4. Color-Based Segmentation - Rice Grains Challenge HSV

7. Appendixes

• Appendix A - Morphology Operators

Instance Segmentation

Not available yet

Semantic Segmentation

1.1. U-Net - Introduction

1.2.

[1.3. U-Net - Retina Challenge (NOT AVAILABLE YET)]

[1.4. DeepLab - Introduction (NOT AVAILABLE YET)]

1.5. DeepLab - Image Segmentation - MobileNetV2

1.6. DeepLab - Video Segmentation - MobileNetV2

1.7. DeepLab - Image Segmentation - Xception 65

1.8. DeepLab - Video Segmentation - Xception 65

Panoptic Segmentation

1.1. Panoptic Segmentation - Introduction

Collaborators:

Rogerio Chaves

✉️ E-mail: chaves.rogerio@outlook.com

Github: https://github.com/Rogerio-Chaves

References:

[1] Montes, C. A. (n.d.). Practical Computer Vision: Theory & Applications [Slide show]. basque center for applied mathematics. http://www.bcamath.org/documentos_public/courses/course_day2.pdf