TRAFFIC SIGNAL RECOGNITION ON AN IMAGE USING CONVOLUTIONAL NEURAL NETWORK
Section: INFORMATIONAL MEASURING SYSTEMS
Authors:
1.
Kazan National Research Technical University named after A.N. Tupolev
2.
Kazan National Research Technical University named after A.N. Tupolev
Type:
Сonference article
Published:
29.12.2021
Subject area:
UDK 62 Инженерное дело. Техника в целом. Транспорт
GRNTI 47.61 Приборы для радиотехнических измерений
GRNTI 47.63 Системы и устройства отображения информации
BBK 302 Проектирование
GRNTI 47.61 Приборы для радиотехнических измерений
GRNTI 47.63 Системы и устройства отображения информации
BBK 302 Проектирование
Language:
Russian
Keywords:
computer vision, classification, neural networks, object recognition, object detection
Abstract (English):
The purpose of this work is to develop an intelligent system for recognizing traffic signals. To achieve this, DetectNet was applied, using an interface for learning, which was developed by NVIDIA. With their help, the disadvantages of this approach were identified, and therefore it was necessary to consider another option for solving this problem.
The purpose of this work is to develop an intelligent system for recognizing traffic signals. To achieve this, DetectNet was applied, using an interface for learning, which was developed by NVIDIA. With their help, the disadvantages of this approach were identified, and therefore it was necessary to consider another option for solving this problem.
Keywords:
computer vision, classification, neural networks, object recognition, object detection
computer vision, classification, neural networks, object recognition, object detection
1. NVIDIA DRIVE - Software: [Elektronnyj resurs] URL: https://developer.nvidia.com/drive/drive-software#driveworks (Data obrashcheniya: 10.10.2020)
2. NVIDIA Developer Blog: [Elektronnyj resurs] URL: https://developer.nvidia.com/blog/detectnet-deep-neural-network-object-detection-digits/ (Data obrashcheniya: 10.10.2020)
3. Pramod Singh, Avinash Manure. Learn TensorFlow 2.0. Apress. 2020.
