Target Detection Based on Simulated Image Domain Migration

Annotating a large amount of data manually for supervised learning is an indispensable and expensive part. A novel system using the simulation dataset is proposed in this paper. This framework can train the neural networks for remote sensing object detection without any manually labeled dataset. The...

Full description

Saved in:
Bibliographic Details
Published in:IEEE access 2020, Vol.8, p.79724-79733
Main Authors: Wang, Yaoling, Gu, Jun, Zhao, Liangjin, Zhang, Yue, Wang, Hongqi
Format: Article
Language:English
Subjects:
Adaptation
Adaptation models
Aircraft
Atmospheric modeling
data simulation
Datasets
Domains
generative and adversarial network
Generators
Image quality
Image reconstruction
Neural networks
Object detection
Object recognition
Remote sensing
Remote sensing image
Simulation
Target detection
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Annotating a large amount of data manually for supervised learning is an indispensable and expensive part. A novel system using the simulation dataset is proposed in this paper. This framework can train the neural networks for remote sensing object detection without any manually labeled dataset. The whole system can be divided into three parts. The first part is the dataset simulator. The simulator synthesizes remote sensing images with the aircraft targets based on real remote sensing images (without any aircraft targets). In the process of data generation, the simulator automatically marks the position information of the aircraft. The second part is the image dataset domain adaptation work. We introduce the work of Cycle-GAN into this part to bridge the perceptual gap between the simulation dataset and reality dataset. Specially, we propose a multi-scale generator into the original Cycle-GAN model to achieve better domain adaptation performance. The final part is the object detection neural network. The domain adaptation quality of the remote sensing images reconstructed by our novel cycle-gan network achieves better performance both in the structural similarity measure and visual appearance. The object detection model trained with the dataset processed by our novel system can get better detection precision. The analytic experiments on the test dataset demonstrate that the object detection model trained with the dataset processed by our novel system can get better detection precision.
ISSN:2169-3536
2169-3536
DOI:10.1109/ACCESS.2020.2989458