Deep transfer learning for underwater vehicle wake recognition in infrared imagery

Zhong Rui; Yang Li; Yongcheng Du

doi:10.1117/12.2544025

18 December 2019 Deep transfer learning for underwater vehicle wake recognition in infrared imagery

Zhong Rui, Yang Li, Yongcheng Du

Proceedings Volume 11333, AOPC 2019: Advanced Laser Materials and Laser Technology; 113330U (2019) https://doi.org/10.1117/12.2544025
Event: Applied Optics and Photonics China (AOPC2019), 2019, Beijing, China

Abstract

With the decreasing noise level of underwater vehicle, the infrared imaging characteristics of underwater vehicle wake become one of its main detectable sources. Using the infrared characteristics of underwater vehicle wake to remote sensing detect the traces of underwater vehicle has gradually developed into a new detection method. Because of the high contingency and large error in judging underwater vehicle wake artificially, it can be overcome by using deep transfer learning to identify and locate the wake. This paper focuses on the infrared feature recognition of underwater vehicle wake with deep transfer learning, and wake sample sets of different classes are produced by image classification. The training effect of different pre-training networks is compared by using deep transfer learning. The influence of internal parameters of pre-training networks on the training effect of wake is discussed. Finally, combined with Faster-RCNN algorithm, the identification effect of wake is tested. The final recognition accuracy is ideal. It has certain application potential for future research on wake remote sensing detection combined with convolution neural network identification.

Citation Download Citation

Zhong Rui, Yang Li, and Yongcheng Du "Deep transfer learning for underwater vehicle wake recognition in infrared imagery", Proc. SPIE 11333, AOPC 2019: Advanced Laser Materials and Laser Technology, 113330U (18 December 2019); https://doi.org/10.1117/12.2544025

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available