Publish In
International Journal of Advance Computational Engineering and Networking (IJACEN)-IJACEN
Journal Home
Volume Issue
Volume-9,Issue-3  ( Mar, 2021 )
Paper Title
CNN as Traffic Sings and Vehicle Classification Model: Model Analysis and Optimizing based on Tensorflow
Author Name
Malicehnkoviktor, Han Honggui
Department of Information, Beijing University of Technology, China
Object detection and classification are among the most investigated types of projects and applicable in a wide range of jobs. One of the main goal is to create such a training model, which contains state-of-the-art technologies and methods with well-tuned parameters, and can be used to scale the project and implement it in real time. This will make it possible, with a luck of a big team, to create an unmanned vehicle that can recognize obstacles and take the necessary actions to avoid accidents. An essential criterion for creating such a model is to use approaches, with a clear analysis (using Tensorboard analysis) and understanding of the significance of each parameter. So we create a unique model that will be as stress-resistant as possible. Proposed article is a part of research of implementing artificial intelligence in different directions of the industry, developed in the laboratory of BJUT. Keywords - CNN, Object Classification, Tensorflow, Tensorboard, Traffic Sings - Vehicles
  View Paper