Identifying Cetacean Mammals in High-Resolution Optical Imagery Using Anomaly Detection Approach Employing Machine Learning Models

• Published in: Moscow University physics bulletin Vol. 78; no. Suppl 1; pp. S149 - S156
• Main Authors: Khabutdinov, I. A., Krinitskiy, M. A., Belikov, R. A.
• Format: Journal Article
• Language: English
• Published: Moscow Pleiades Publishing 01-12-2023; Springer Nature B.V
• Subjects: Anomalies; Artificial neural networks; Counting; Datasets; Dolphins; Dolphins & porpoises; Expeditions; Helicopters; High resolution; Image reconstruction; Image resolution; Machine learning; Machine Learning in Natural Sciences; Mammals; Mathematical and Computational Physics; Monitoring; Physics; Physics and Astronomy; Populations; Spatial distribution; Theoretical; Black Sea; deep learning; artificial neural networks; U-Net; convolutional autoencoder; computer vision; anomaly detection; machine learning; cetacean mammals
• ISSN: 0027-1349; 1934-8460
• DOI: 10.3103/S0027134923070147

Cetacean mammal populations, particularly dolphins, have recently experienced significant declines due to various artificial and natural factors. A crucial aspect of studying these populations is determining their numbers and assessing spatial distributions. In our study, we focus on monitoring dolphin populations in the Black Sea using high-resolution photographs taken from helicopters for counting purposes. Currently, expert analysts manually count dolphins in these images, which is a time-consuming process. To address this issue, we propose the use of machine learning (ML) approaches, specifically, anomaly detection using ML models. We examine a dataset collected during accounting marine expeditions of the Shirshov Institute of Oceanology of the Russian Academy of Sciences (IORAS) in the Black Sea from 2018 to 2019. The dataset consists of 3730 high-resolution photographs, with dolphins present in 205 images (5.5
). Each dolphin occupies approximately 0.005
of an image area (around
pixels), making their presence a rare event. Thus, we treat dolphin identification as an anomaly detection task. Our study compares classical and naive anomaly detection methods with reconstruction-based approaches that discriminate anomalies based on the magnitude of reconstruction errors. Within this latter approach, we utilize various artificial neural networks, such as Convolutional Autoencoders (CAE) and U-Net, for image reconstruction. Overall, our research aims to streamline the process of counting and monitoring dolphin populations in high-resolution imagery using advanced ML techniques.