Learning Object Classification

In Chapter 5, Automated Optical Inspection, Object Segmentation, and Detection, we introduced the basic concepts of object segmentation and detection. This refers to isolating the objects that appear in an image for future processing and analysis. This chapter explains how to classify each of these isolated objects. To allow us to classify each object, we have to train our system to be capable of learning the required parameters so that it decides which specific label will be assigned to the detected object (depending on the different categories taken into account during the training phase).

This chapter introduces the basics concepts of machine learning to classify images with different labels. To do this, we are going to create a basic application based on the segmentation algorithm of Chapter 5, Automated Optical Inspection, Object Segmentation, and Detection. This segmentation algorithm extracts parts of images that contain unknown objects. For each detected object, we are going to extract different features that are going to be classified using a machine learning algorithm. Finally, we are going to show the obtained results using our user interface, together with the labels of each object detected in the input image.

This chapter involves different topics and algorithms, including the following:

• Introduction to machine learning concepts
• Common machine learning algorithms and processes
• Feature extraction
• support vector machines (SVM)
• Training and prediction