Aquatic ecologists routinely count animals to provide critical information for conservation and management. Increased accessibility to underwater recording equipment such as action cameras and unmanned underwater devices has allowed footage to be captured efficiently and safely, without the logistical difficulties manual data collection often presents. It has, however, led to immense volumes of data being collected that require manual processing and thus significant time, labor, and money. The use of deep learning to automate image processing has substantial benefits but has rarely been adopted within the field of aquatic ecology. To test its efficacy and utility, we compared the accuracy and speed of deep learning techniques against human counterparts for quantifying fish abundance in underwater images and video footage. We collected footage of fish assemblages in seagrass meadows in Queensland, Australia. We produced three models using an object detection framework to detect the target species, an ecologically important fish, luderick (Girella tricuspidata). Our models were trained on three randomized 80:20 ratios of training:validation datasets from a total of 6,080 annotations. The computer accurately determined abundance from videos with high performance using unseen footage from the same estuary as the training data (F1 = 92.4%, mAP50 = 92.5%) and from novel footage collected from a different estuary (F1 = 92.3%, mAP50 = 93.4%). The computer's performance in determining abundance was 7.1% better than human marine experts and 13.4% better than citizen scientists in single image test datasets, and 1.5 and 7.8% higher in video datasets, respectively. We show that deep learning can be a more accurate tool than humans at determining abundance and that results are consistent and transferable across survey locations. Deep learning methods provide a faster, cheaper, and more accurate alternative to manual data analysis methods currently used to monitor and assess animal abundance and have much to offer the field of aquatic ecology.

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

No abstract

Environmental monitoring guides conservation and is particularly important for aquatic habitats which are heavily impacted by human activities. Underwater cameras and uncrewed devices monitor aquatic wildlife, but manual processing of footage is a significant bottleneck to rapid data processing and dissemination of results. Deep learning has emerged as a solution, but its ability to accurately detect animals across habitat types and locations is largely untested for coastal environments. Here, we produce five deep learning models using an object detection framework to detect an ecologically important fish, luderick (Girella tricuspidata). We trained two models on footage from single habitats (seagrass or reef) and three on footage from both habitats. All models were subjected to tests from both habitat types. Models performed well on test data from the same habitat type (object detection measure: mAP50: 91.7 and 86.9% performance for seagrass and reef, respectively) but poorly on test sets from a different habitat type (73.3 and 58.4%, respectively). The model trained on a combination of both habitats produced the highest object detection results for both tests (an average of 92.4 and 87.8%, respectively). The ability of the combination trained models to correctly estimate the ecological abundance metric, MaxN, showed similar patterns. The findings demonstrate that deep learning models extract ecologically useful information from video footage accurately and consistently and can perform across habitat types when trained on footage from the variety of habitat types.

No abstract