Foreground enhancement network for object detection in sonar images

As a special detection task, sonar image object detection has been suffering from two main problems: the widespread noise and the lack of high-frequency information. In this paper, we propose two independent modules to solve the above two problems. For the widespread noise, we propose the foreground semantic enhancement module. Different from simple feature fusion, this module creatively associates the semantic map with features from each feature level, thus increasing the foreground-background distance and highlighting the object information. To solve the problem of insufficient high-frequency information, we propose the foreground edge enhancement module. This module inventively combines RNN networks to enhance edges by spatial semantic information from different directions as a way to improve the feature representation of foreground objects. Based on the above two modules, we design a novel detection architecture, foreground enhancement network (FEN), which enhances the features of a single point to make the classification more powerful and the localization more accurate. Through extensive experimental validation, our FEN network achieves high-performance improvement when combined with different detectors, and achieves the highest 10% mAP performance improvement when combined with a single-stage detector (FCOS).

Automated summary

In this paper, two modules are proposed to solve the problems of widespread noise and lack of high-frequency information in sonar image object detection. The foreground semantic enhancement module associates the semantic map with features to increase the foreground-background distance, while the foreground edge enhancement module enhances edges by spatial semantic information. A novel detection architecture, FEN network, is designed based on these modules to improve classification and localization accuracy.