TRAFFIC SIGN RECOGNITION METHOD USING DEEP NEURAL NETWORK BASED ON ONLINE UNSUPERVISED LEARNING AND SYSTEM THEREOF

摘要

Provided are a traffic sign recognition method using a deep neural network based on online unsupervised learning and a system thereof. According to an embodiment of the present invention, an object recognition system comprises: an extractor extracting a feature of an object using a neural network based on unsupervised learning; and a classifier classifying the detected object based on the extracted feature. Accordingly, since the object such as a traffic signal is recognized by using a deep neural network based on online unsupervised learning, speed online learning at an edge device stage becomes possible.;COPYRIGHT KIPO 2019