Search

To simulate the underwater environment and test algorithms for autonomous underwater vehicles, we developed an underwater simulation environment with the Unreal Engine 4 to generate underwater visual data such as seagrass and landscape. We then used such data from the Unreal environment to train and verify an underwater image segmentation model, which is an important technology to later achieve visual based navigation. The simulation environment shows the potentials for dataset generalization and testing robot vision algorithms.

A deep neural network architecture is proposed in this paper for underwater scene semantic segmentation. The architecture consists of encoder and decoder networks. Pretrained VGG-16 network is used as a feature extractor, while the decoder learns to expand the lower resolution feature maps. The network applies max un-pooling operator to avoid large number of learnable parameters, and, in order to make use of the feature maps in encoder network, it concatenates the feature maps with decoder and encoder for lower resolution feature maps. Our architecture shows capabilities of faster convergence and better accuracy. To get a clear view of underwater scene, an underwater enhancement neural network architecture is described in this paper and applied for training. It speeds up the training process and convergence rate in training.