Machine learning and neural networks are now ubiquitous in sonar perception, but it lags behind the computer vision field due to the lack of data and pre-trained models specifically for sonar images. In this paper we present the Marine Debris Turntable dataset and produce pre-trained neural networks trained on this dataset, meant to fill the gap of missing pre-trained models for sonar images. We train Resnet 20, MobileNets, DenseNet121, SqueezeNet, MiniXception, and an Autoencoder, over several input image sizes, from 32 x 32 to 96 x 96, on the Marine Debris turntable dataset. We evaluate these models using transfer learning for low-shot classification in the Marine Debris Watertank and another dataset captured using a Gemini 720i sonar. Our results show that in both datasets the pre-trained models produce good features that allow good classification accuracy with low samples (10-30 samples per class). The Gemini dataset validates that the features transfer to other kinds of sonar sensors. We expect that the community benefits from the public release of our pre-trained models and the turntable dataset.

本文介绍了Marine Debris Turntable数据集和基于此数据集训练的预训练神经网络，可用于填补声纳图像缺失的预训练模型的空白。通过使用转移学习，在Marine Debris Watertank和Gemini 720i sonar数据集上对训练的模型进行评估，结果表明在低样本（每类10-30个样本）下，预训练的模型能够产生良好的特征，提高分类准确性，并且这些特征能够转移到到其他类型的声纳传感器中。

用于声纳图像的预训练模型