Automated Optical Inspection, Object Segmentation, and Detection

In Chapter 4, Delving into Histogram and Filters, we learned about histograms and filters, which allow us to understand image manipulation and create a photo application.

In this chapter, we are going to introduce the basic concepts of object segmentation and detection. This means isolating the objects that appear in an image for future processing and analysis.

This chapter introduces the following topics:

• Noise removal
• Light/background removal basics
• Thresholding
• Connected components for object segmentation
• Finding contours for object segmentation

Many industries use complex computer vision systems and hardware. Computer vision tries to detect problems and minimize errors produced in the production process, improving the quality of final products.

In this sector, the name for this computer vision task is Automated Optical Inspection (AOI). This name appears in the inspection of printed circuit board manufacturers, where one or more cameras scan each circuit to detect critical failures and quality defects. This nomenclature was used in other manufacturing industries so that they could use optical camera systems and computer vision algorithms to increase product quality. Nowadays, optical inspection using different camera types (infrared or 3D cameras), depending on the requirements, and complex algorithms are used in thousands of industries for different purposes such as defect detection, classification, and so on.