Unsupervised Knowledge Transfer for Object Detection in Marine Environmental Monitoring and Exploration

Abstract

The volume of digital image data collected in the field of marine environmental monitoring and exploration has been growing in rapidly increasing rates in recent years. Computational support is essential for the timely evaluation of the high volume of marine imaging data, but often modern techniques such as deep learning cannot be applied due to the lack of training data. In this paper, we present Unsupervised Knowledge Transfer (UnKnoT), a new method to use the limited amount of training data more efficiently. In order to avoid time-consuming annotation, it employs a technique we call "scale transfer" and enhanced data augmentation to reuse existing training data for object detection of the same object classes in new image datasets. We introduce four fully annotated marine image datasets acquired in the same geographical area but with different gear and distance to the sea floor. We evaluate the new method on the four datasets and show that it can greatly improve the object detection performance in the relevant cases compared to object detection without knowledge transfer. We conclude with a recommendation for an image acquisition and annotation scheme that ensures a good applicability of modern machine learning methods in the field of marine environmental monitoring and exploration.

Guiding Labelling Effort for Efficient Learning With Georeferenced Images

“…Furthermore, for sub-sea imaging, most groups gather images using custom built imaging hardware, where in [28] the authors reported that even small differences in sub-sea imaging hardware limits learning transferability and distorts deep learning classifier outputs. In [29] a pipeline to make training datasets transferable for inference on images from other datasets is proposed for segmentation of marine organism. The work proposes how to reduce scale variance across multiple datasets, which is highlighted as an important consideration for seafloor imagery.…”

Computer vision and deep learning for fish classification in underwater habitats: A survey

“…A well-known and effective method for improving the generalizability of a DL model is to use regularization (Kukacˇka et al, 2017). Some of the regularization methods applied to fish and marine habitat monitoring domains include transfer learning (Zurowietz & Nattkemper, 2020), batch normalization (Islam et al, 2020), dropout (Iqbal et al, 2021) and using a regularization term (Tarling et al, 2021).…”

Computer Vision and Deep Learning for Fish Classification in Underwater Habitats: A Survey

“…A well-known and effective method for improving the generalisability of a DL model is to use regularisation [87]. Some of the regularisation methods applied to fish and marine habitat monitoring domains include transfer learning [88], batch normalisation [57], dropout [81], and using a regularisation term [62]. Since it is difficult to obtain a large labelled dataset, various techniques have been proposed to address this challenge.…”