Application of underwater robots are on the rise, most of them are dependent on sonar for underwater vision, but the lack of strong perception capabilities limits them in this task. An important issue in sonar perception is matching image patches, which can enable other techniques like localization, change detection, and mapping. There is a rich literature for this problem in color images, but for acoustic images, it is lacking, due to the physics that produce these images. In this paper we improve on our previous results for this problem (Valdenegro-Toro et al, 2017), instead of modeling features manually, a Convolutional Neural Network (CNN) learns a similarity function and predicts if two input sonar images are similar or not. With the objective of improving the sonar image matching problem further, three state of the art CNN architectures are evaluated on the Marine Debris dataset, namely DenseNet, and VGG, with a siamese or two-channel architecture, and contrastive loss. To ensure a fair evaluation of each network, thorough hyper-parameter optimization is executed. We find that the best performing models are DenseNet Two-Channel network with 0.955 AUC, VGG-Siamese with contrastive loss at 0.949 AUC and DenseNet Siamese with 0.921 AUC. By ensembling the top performing DenseNet two-channel and DenseNet-Siamese models overall highest prediction accuracy obtained is 0.978 AUC, showing a large improvement over the 0.91 AUC in the state of the art.
翻译:水下机器人的应用呈上升趋势, 大部分都依赖于声纳, 但缺乏强大的感知能力限制了他们完成这一任务。 声纳感知的一个重要问题是匹配图像补丁, 这可以促进其它技术, 如本地化、 变化检测和绘图等。 在彩色图像中, 这个问题有丰富的文献, 但是由于产生这些图像的物理原理, 声学图像却缺乏。 在本文中, 我们改进了我们以前对这一问题的结果( Valdenegro- Toro等人, 2017年), 而不是人工建模功能, 一个革命性神经网络( CNN) 学习了相似的功能, 并且预测了两种输入声纳图像是否相似。 为了进一步改进声纳图像匹配问题, 在海洋碎片数据集上, 即DesenseNet, VGG, 有三个状态的艺术CNN架构, 以及一个感光或两层气流结构, 以确保对每个网络进行公平的评估, 彻底的超度校校准。 我们发现, 运行的模型是最高值的 AS- AS- AS- AS- AL AS- AL AL ASy ASy ASU A. AS- AS- 0.55 和整个 ASy ASy ASy ASU ASyal AS- AS- ASU ASyal AS- AS- AS- AS- AS- AS- AS AS- AS- AS- AS- sli 0. AS- AS- sli AL 0. ASU ASy 0. ASU 0. AL 0. AS AS AS ASU 1 AS AS AS 0. AS AS AS AS AS AS AS AS AS AS AS AS AS AS AS AS AS AS AS AS AL AL AL AS AS AS- AS- ASU BAR AS- 0. AS AS AS AS AS AS AS 0. 0. AS AS AL AL AS- AS- AS- AS- AS- 0. 0. 0. 0. 0. AS- AS- AL AS