基于地面回波建模的二维前视声呐仿真
2D Forward Looking Sonar Simulation with Ground Echo Modeling
AI总结：
本文针对二维前视声呐仿真中多径反射建模不足的问题，提出一种基于单次反射框架物理仿真平坦地面回波效应的方法，并通过水池真实图像对比验证了其有效性。
摘要：
成像声呐在水下环境中能够产生清晰的图像，不受水体浑浊度和光照条件的影响。新一代二维前视声呐体积紧凑，能够生成高分辨率图像，从而促进了水下机器人研究。考虑到在水下环境中开展实验的困难和费用，大量工作集中在声呐图像仿真上。然而，多径反射等声呐伪影尚未得到充分讨论，而这些伪影在水池环境中不可忽视。本文重点研究来自平坦地面的回波影响。我们提出了一种在声学图像中物理仿真地面回波效应的方法。为了提高计算效率，我们使用单次反射框架对多次反射情况进行建模。我们将水池中采集的真实图像与合成图像进行对比，以验证所提方法的有效性。
英文摘要：
Imaging sonar produces clear images in underwater environments, independent of water turbidity and lighting conditions. The next generation 2D forward looking sonars are compact in size and able to generate high-resolution images which facilitate underwater robotics research. Considering the difficulties and expenses of implementing experiments in underwater environments, tremendous work has been focused on sonar image simulation. However, sonar artifacts like multi-path reflection were not sufficiently discussed, which cannot be ignored in water tank environments. In this paper, we focus on the influence of echoes from the flat ground. We propose a method to simulate the ground echo effect physically in acoustic images. We model the multi-bounce situations using the single-bounce framework for computation efficiency. We compare the real image captured in the water tank with the synthetic images to validate the proposed methods.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
