A water-obstacle separation and refinement network for unmanned surface vehicles
This addresses obstacle detection for autonomous navigation in unmanned surface vehicles, representing an incremental improvement over existing methods.
The paper tackles the problem of poor water edge estimation, small obstacle detection, and high false positives in obstacle detection for unmanned surface vehicles by proposing a new deep encoder-decoder architecture called WaSR, which results in a 14% increase in F-measure over the state-of-the-art.
Obstacle detection by semantic segmentation shows a great promise for autonomous navigation in unmanned surface vehicles (USV). However, existing methods suffer from poor estimation of the water edge in the presence of visual ambiguities, poor detection of small obstacles and high false-positive rate on water reflections and wakes. We propose a new deep encoder-decoder architecture, a water-obstacle separation and refinement network (WaSR), to address these issues. Detection and water edge accuracy are improved by a novel decoder that gradually fuses inertial information from IMU with the visual features from the encoder. In addition, a novel loss function is designed to increase the separation between water and obstacle features early on in the network. Subsequently, the capacity of the remaining layers in the decoder is better utilised, leading to a significant reduction in false positives and increased true positives. Experimental results show that WaSR outperforms the current state-of-the-art by a large margin, yielding a 14% increase in F-measure over the second-best method.