In this paper, we propose automatic road extraction using Unmanned Aerial Vehicle (UAV) based Remote Sensing data. Road extraction using UAV data is very useful in traffic management, city planning, GPS based applications, etc. Deep learning techniques namely, Fully Convolutional Network (FCN) and conditional Generative Adversarial Networks (GAN) are used to extract roads from a UAV dataset available in the literature. FCN performs semantic segmentation on the image whereas the GAN generates output images from the model it learns. The results demonstrate the efficiency of the deep learning methods for the task of road extraction.
DeepExt: A Convolution Neural Network for Road Extraction using RGB images captured by UAV
01.11.2018
314159 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
SODCNN: A Convolutional Neural Network Model for Small Object Detection in Drone-Captured Images
DOAJ | 2023
|Road section traffic flow prediction method based on time convolution neural network
Europäisches Patentamt | 2020
|