University of Leicester
Browse
sun-2017.pdf (1023.74 kB)

Transferring deep knowledge for object recognition in Low-quality underwater videos

Download (1023.74 kB)
journal contribution
posted on 2018-01-30, 14:55 authored by Xin Sun, Junyu Shi, Lipeng Liu, Junyu Dong, Claudia Plant, Xinhua Wang, Huiyu Zhou
In recent years, underwater video technologies allow us to explore the ocean in scientific and noninvasive ways, such as environmental monitoring, marine ecology studies, and fisheries management. However the low-light and high-noise scenarios pose great challenges for the underwater image and video analysis. We here propose a CNN knowledge transfer framework for underwater object recognition and tackle the problem of extracting discriminative features from relatively low contrast images. Even with the insufficient training set, the transfer framework can well learn a recognition model for the special underwater object recognition task together with the help of data augmentation. For better identifying objects from an underwater video, a weighted probabilities decision mechanism is introduced to identify the object from a series of frames. The proposed framework can be implemented for real-time underwater object recognition on autonomous underwater vehicles and video monitoring systems. To verify the effectiveness of our method, experiments on a public dataset are carried out. The results show that the proposed method achieves promising results for underwater object recognition on both test image datasets and underwater videos.

Funding

This work is supported by the National Natural Science Foundation of China (No. 61401413, 41576011), China Postdoctoral Science Foundation (No. 2015T80749), Natural Science Foundation of Shandong Province (No. ZR2014FQ023), Open Funding of State Key Laboratory of Applied Optics and NVIDIA Academic Hardware Grant.

History

Citation

Neurocomputing, 2017, 275, pp. 897-908

Author affiliation

/Organisation

Version

  • AM (Accepted Manuscript)

Published in

Neurocomputing

Publisher

Elsevier

issn

0925-2312

Acceptance date

2017-09-12

Copyright date

2017

Available date

2018-09-21

Publisher version

https://www.sciencedirect.com/science/article/pii/S0925231217315631?via=ihub

Notes

The file associated with this record is under embargo until 12 months after publication, in accordance with the publisher's self-archiving policy. The full text may be available through the publisher links provided above.

Language

en

Usage metrics

    University of Leicester Publications

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC