After the growth of AI and machine learning, machines can recognize handwriting and images. Those skills are widely used in real life; for example, one of the IPad's features is recognizing the user's notes and turning them into text. The most fundamental of those recognitions is digit recognition. Digit recognition is a process of identifying handwritten digits from 0 to 9. 

This technology has several applications: automatic license plate recognition, postal sorting, and online banking. Automatic license plate recognition works with this technology to read the digits on the license plate, and it eventually helps the police to recognize the vehicles in interest. Due to the significant improvements in deep learning, digit recognition accuracy is growing, and there are several solutions and algorithms to identify the digits. 

The most prominent neural network is CNN (Convolutional Neural Network). Before getting familiar with CNN, it is essential to understand the basics of neural networks. Neural networks are the foundation of machine learning and expanding the fields of artificial intelligence. The name neural networks were inspired by how the human brain sends signals to communicate with individual components. 

One neuron receives input from another and sends the output; if the result (weight) is over the threshold, the function is activated. The threshold is based on trained data to change the neurons' behavior over time. This is why it is called machine learning, where the neurons learn new information as human brains and come to a result. CNN was developed based on this concept. 

이주예 강남포스트 학생기자  webmaster@ignnews.kr

<저작권자 © 강남포스트, 무단 전재 및 재배포 금지>