Underwater fish detection with weak multi-domain supervision

Konovalov, Dmitry A., Saleh, Alzayat, Bradley, Michael, Sankupellay, Mangalam, Marini, Simone, and Sheaves, Marcus (2019) Underwater fish detection with weak multi-domain supervision. In: Proceedings of the International Joint Conference on Neural Networks. From: 2019 IJCNN: International Joint Conference on Neural Networks, 14-19 July 2019, Budapest, Hungary.

[img] PDF (Published Version) - Published Version
Restricted to Repository staff only

View at Publisher Website: https://doi.org/10.1109/IJCNN.2019.88519...
 
2


Abstract

Given a sufficiently large training dataset, it is relatively easy to train a modern convolution neural network (CNN) as a required image classifier. However, for the task of fish classification and/or fish detection, if a CNN was trained to detect or classify particular fish species in particular background habitats, the same CNN exhibits much lower accuracy when applied to new/unseen fish species and/or fish habitats. Therefore, in practice, the CNN needs to be continuously fine-tuned to improve its classification accuracy to handle new project-specific fish species or habitats. In this work we present a labelling- efficient method of training a CNN-based fish-detector (the Xception CNN was used as the base) on relatively small numbers (4,000) of project-domain underwater fish/no-fish images from 20 different habitats. Additionally, 17,000 of known negative (that is, missing fish) general-domain (VOC2012) above-water images were used. Two publicly available fish-domain datasets supplied additional 27,000 of above-water and underwater positive/fish im- ages. By using this multi-domain collection of images, the trained Xception-based binary (fish/not-fish) classifier achieved 0.17% false-positives and 0.61% false-negatives on the project’s 20,000 negative and 16,000 positive holdout test images, respectively. The area under the ROC curve (AUC) was 99.94%.

Item ID: 60793
Item Type: Conference Item (Research - E1)
ISBN: 978-1-7281-1985-4
ISSN: 2161-4407
Related URLs:
Date Deposited: 03 Dec 2019 00:25
FoR Codes: 08 INFORMATION AND COMPUTING SCIENCES > 0801 Artificial Intelligence and Image Processing > 080106 Image Processing @ 100%
SEO Codes: 96 ENVIRONMENT > 9608 Flora, Fauna and Biodiversity > 960808 Marine Flora, Fauna and Biodiversity @ 100%
Downloads: Total: 2
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page