Chapter 16. Number Plate Recognition with Deep Convolutional Networks
This chapter introduces us to the steps needed to create an application for Automatic Number Plate Recognition (ANPR). There are different approaches and techniques based on different situations; for example, an infrared camera, fixed car position, and light conditions. We can proceed to construct an ANPR application to detect automobile license plates in a photograph taken between two and three meters from a car, in ambiguous light conditions, and with a non-parallel ground with minor perspective distortions in the automobile's plate.
The main purpose of this chapter is to introduce us to image segmentation and feature extraction, pattern recognition basics, and two important pattern recognition algorithms, the Support Vector Machine (SVM) and deep neural network (DNN), using convolutional networks. In this chapter, we will cover the following topics:
- ANPR
- Plate detection
- Plate recognition
