Here we will highlight some of the past and ongoing research utilizing LoVe Ocean data, starting with a brief overview of the research connected with images and monitoring of cold-water coral structures at Node 1.
Cold-water corals (PI: Ingunn Nilssen)
The longest line of research with LoVe Ocean data to date focuses on the images acquired by the Node 1 Satellites. Before the new nodes (Nodes 2-7), there were two locations delivering images, camera 1 (Node 1 - Satellite 1) where a HD camera is facing a cold-water coral structure of Desmophyllum pertusum (previous Lophelia pertusa), and camera 2 (Node 2 - Satellite 2) where a stereo camera is facing bubble-gum corals (Peragorgia arborea). The latter also alter between two positions, one that take a close-up of a bubble-gum coral and one position covering several bubble-gum coral individuals on a distance.
Till present, all work has been focused on photos from the cold-water coral structure and associated organisms. Example images can be found in the Gallery section of loveocean.no. Images from LoVe have been used for testing and development of algorithms to detect various organisms’ behaviour. Further, converting the images to numerical values has enabled the investigation of natural variations, and to what extent the various measured physical or chemical parameters influence behaviour.
A variety of publications, focusing either on specific species or screening using change detection and classification, have been published in peer-reviewed journals. These papers are listed below and in the LoVe Ocean publication list.
Osterloff, J. et al. 2016a. Computational visual stress level analysis of calcareous algae exposed to sedimentation. PLoS ONE Vol 11, pp e0157329, doi: http://dx.doi.org/10.1371/journal.pone.0157329.
Osterloff, J. et al. 2016b. A computer vision approach for monitoring the spatial and temporal shrimp distribution at the LoVe observatory. Methods in Oceanography, Vol 15-16, pp 114-128, doi: http://dx.doi.org/10.1016/j.mio.2016.03.002.
Osterloff, J. et al. 2016c. Computational coral feature monitoring for the fixed underwater observatory LoVe. OCEANS 2016 MTS/IEEE Monterey, Monterey, CA, pp. 1-5.
doi: http://dx.doi.org/10.1109/OCEANS.2016.7761417
Osterloff, J. et al. 2016d. Polyp Activity Estimation and Monitoring for Cold Water Corals with a Deep Learning Approach. 2016 ICPR 2nd Workshop on Computer Vision for Analysis of Underwater Imagery (CVAUI), Cancun, pp 1-6.
doi: http://dx.doi.org/10.1109/CVAUI.2016.013
Osterloff, J. et al. 2019. Computer vision enables short and long-term analysis of Lophelia pertusa polyp behaviour and colour from an underwater observatory. Scientific reports 9, 6578 (2019). doi: https://www.nature.com/articles/s41598-019-41275-1
Möller, T. et al. 2016a. Change detection in marine observatory image streams using Bi-Domain Feature Clustering. 23rd International Conference on Pattern Recognition (ICPR). pp 793-798.
doi: http://dx.doi.org/10.1109/ICPR.2016.7899732
Möller, T. et al. 2016b. Data-driven long term change analysis in marine observatory image streams. ICPR 2nd Workshop on Computer Vision for Analysis of Underwater Imagery (CVAUI), Cancun, pp. 13-18.
doi: http://dx.doi.org/10.1109/CVAUI.2016.015
Möller, T. et al. 2017. Active Learning for the Classification of Species in Underwater Images from a Fixed Observatory. IEEE International Conference on Computer Vision Workshops (ICCVW), Venice, pp. 2891-2897.
doi: http://dx.doi.org/10.1109/ICCVW.2017.341
Möller T. et.al. 2019. Tracking Sponge Size and Behaviour with Fixed Underwater Observatories. In: Zhang Z., Suter D., Tian Y., Branzan Albu A., Sidère N., Jair Escalante H. (eds) Pattern Recognition and Information Forensics. ICPR 2018. Lecture Notes in Computer Science, vol 11188. Springer, Cham
doi: https://doi.org/10.1007/978-3-030-05792-3_5