Artificial Intelligence Programming with Python

CHAPTER 5Image Classification

“If our era is the next Industrial Revolution, as many claim, AI is surely one of its driving forces.”

—Fei-Fei Li (American computer scientist)

CHAPTER OUTLINE

1. 5.1. Introduction
2. 5.2. Classifications with Pre-trained Models
3. 5.3. Classifications with Custom Trained Models: Transfer Learning
4. 5.4. Cancer/Disease Detections
5. 5.5. Federated Learning for Image Classification
6. 5.6. Web-Based Image Classification
7. 5.7. Image Processing
8. 5.8. Summary
9. 5.9. Chapter Review Questions

5.1 Introduction

Image analysis or image processing is the most researched topic in deep learning. Image processing simply means extracting information from images and learning from images. This is an important aspect of computer vision. In deep learning, image processing is largely done using convolutional neural networks (CNNs) and can be generally divided into image classification, object detection, image segmentation, and so on.

Figure 5.1 illustrates the differences between image classification, object detection, and image segmentation. Image classification is about identifying the image or classifying the image. Image classification can be typically done by using deep learning neural networks such as AlexNet, GoogLeNet, VGG, ResNet, MobileNet, and so on. Object detection is about identifying a particular object in the image. Object detection can be typically done by using deep learning neural networks such as region-based convolutional neural networks (R-CNNs), you ...