Abstract Object detection using deep learning over the years became one of the most popular methods for implementation in autonomous systems. Autonomous vehicle requires very reliable and accurate identification and recognition of surrounding objects in real traffic environments to achieve decent detection results. In this paper, special type of Artificial Neural Network (ANN) named Convolutional Neural Network (CNN) was used for identification and recognition of surrounding objects in real traffic. The new model based on CNN was trained and developed to be able to identify and recognize 4 different classes of objects: cars, traffic lights, persons and bicycles. The developed model has shown 94.6% accuracy of object identification and recognizing on the test set.
Identification and Recognition of Vehicle Environment Using Artificial Neural Networks
04.09.2018
12 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Aircraft System Identification Using Artificial Neural Networks
British Library Conference Proceedings | 2013
|