Deep learning based deep-sea automatic image enhancement and animal species classification

被引:0
作者
Vanesa Lopez-Vazquez
Jose Manuel Lopez-Guede
Damianos Chatzievangelou
Jacopo Aguzzi
机构
[1] University of the Basque Country (UPV/EHU),Department of System Engineering and Automation Control
[2] University of the Basque Country (UPV/EHU),Department of Renewable Marine Resources
[3] Instituto de Ciencias del Mar (ICM-CSIC),undefined
[4] Instituto de Ciencias del Mar (ICM-CSIC),undefined
[5] Stazione Zoologica of Naples (SZN) Anton Dohrn,undefined
来源
Journal of Big Data | / 10卷
关键词
Underwater image enhancement; Deep learning; Neural networks; Image classification;
D O I
暂无
中图分类号
学科分类号
摘要
The automatic classification of marine species based on images is a challenging task for which multiple solutions have been increasingly provided in the past two decades. Oceans are complex ecosystems, difficult to access, and often the images obtained are of low quality. In such cases, animal classification becomes tedious. Therefore, it is often necessary to apply enhancement or pre-processing techniques to the images, before applying classification algorithms. In this work, we propose an image enhancement and classification pipeline that allows automated processing of images from benthic moving platforms. Deep-sea (870 m depth) fauna was targeted in footage taken by the crawler “Wally” (an Internet Operated Vehicle), within the Ocean Network Canada (ONC) area of Barkley Canyon (Vancouver, BC; Canada). The image enhancement process consists mainly of a convolutional residual network, capable of generating enhanced images from a set of raw images. The images generated by the trained convolutional residual network obtained high values in metrics for underwater imagery assessment such as UIQM (~ 2.585) and UCIQE (2.406). The highest SSIM and PSNR values were also obtained when compared to the original dataset. The entire process has shown good classification results on an independent test data set, with an accuracy value of 66.44% and an Area Under the ROC Curve (AUROC) value of 82.91%, which were subsequently improved to 79.44% and 88.64% for accuracy and AUROC respectively. These results obtained with the enhanced images are quite promising and superior to those obtained with the non-enhanced datasets, paving the strategy for the on-board real-time processing of crawler imaging, and outperforming those published in previous papers.
引用
收藏
相关论文
共 461 条
[21]  
Fanelli E(2018)Machine learning approaches for crop yield prediction and nitrogen status estimation in precision agriculture: a review Comput Electron Agric 151 8821-14753
[22]  
Danovaro R(2018)A low-complexity mosaicing algorithm for stock assessment of seabed-burrowing species IEEE J Oceanic Eng 44 452-33177
[23]  
Flögel S(2010)Surface area and the seabed area, volume, depth, slope, and topographic variation for the world’s seas, oceans, and countries Environ Sci Technol 44 181-374
[24]  
Lebris N(2017)A new international ecosystem-based strategy for the global deep ocean Science 355 17995-874
[25]  
Juanes F(2020)Ecological variables for developing a global deep-ocean monitoring and conservation strategy Nat Ecol Evol 4 14740-10
[26]  
De Leo FC(2012)The 27–year decline of coral cover on the Great Barrier Reef and its causes Proc Natl Acad Sci 109 33163-47
[27]  
Del Rio J(2014)A new colorimetrically-calibrated automated video-imaging protocol for day-night fish counting at the OBSEA coastal cabled observatory Sensors 13 358-475
[28]  
Aguzzi J(2020)Obsea: a decadal balance for a cabled observatory deployment IEEE Access 8 861-853
[29]  
Costa C(2017)Seasonal monitoring of deep-sea cold-seep benthic communities using an Internet Operated Vehicle (IOV) PLoS ONE 12 1-123
[30]  
Calisti M(2013)NEMO-SN1 abyssal cabled observatory in the Western Ionian Sea IEEE J Oceanic Eng 38 38-2324