MASNet: A Robust Deep Marine Animal Segmentation Network

Marine animal studies are of great importance to human beings and instrumental to many research areas. How to identify such animals through image processing is a challenging task that leads to marine animal segmentation (MAS). Although deep neural networks have been widely applied for object segmentation, few of them consider the complex imaging condition in the water and the camouflage property of marine animals. To this end, a robust deep marine animal segmentation network is proposed in this article. Specifically, we design a new data augmentation strategy to randomly change the degradation and camouflage attributes of the original objects. With the augmentations, a fusion-based deep neural network constructed in a Siamese manner is trained to learn the shared semantic representations. Moreover, we construct a new large-scale real-world MAS data set for conducting extensive experiments. It consists of over 3000 images with various underwater scenes and objects. Each image is annotated with an object-level mask and assigned to a category. Extensive experimental results show that our method significantly outperforms 12 state-of-the-art methods both qualitatively and quantitatively.

I. Introduction

The exploration of underwater environments has been an active engagement across a plethora of scientific fields, such as ocean ecology, marine geological sciences, and natural resources discovery. Recently, visually guided underwater robots and intelligent underwater monitoring systems become instrumental tools or equipment to assist these activities effectively, and many image processing algorithms have been developed to serve various purposes. However, existing solutions for conducting marine scene parsing and object segmentation are largely underexplored. In this article, it is our great interest in developing image segmentation algorithms to perform marine animal segmentation (MAS), particularly for a highly ill-conditioned underwater environment.