本文为瑞典查尔姆斯理工大学（作者：Emil Andersson）的硕士论文，共34页。

本论文研究在声纳和水下摄像机图像上使用滤波器和先进算法的可能性。用于正常图像的滤波器对水下图像是否同样有用；是否可以将这些滤波器用于表示声音反射而不是光反射的声纳图像？除了普通滤波器（图像增强）外，还将对模式匹配算法进行并行研究。本课题是最近才开始在监控中应用的，其实现和算法都比较先进。研究将发现这些算法是否足够有效，能够实时运行，同时向用户提供有用的数据。作为本文的一部分，将为澳大利亚海关开发水下检查软件。在大多数情况下，这将为滤波器和算法在实际情况下和/或在实际检查的材料上进行测试提供额外的好机会。

The thesis is researching the possibilities to use filters and advanced algorithms on sonar and underwater camera images. Can filters used for normal images be as helpful on underwater images; and is it possible to use these filters on sonar images that represent sound reflection instead of light reflections? Besides the normal filters (image enhancements) a parallel research on pattern matching algorithms will take place. This topic is just recently started to being used in surveillance applications; the implementations and algorithms are advanced. The research will find out if these algorithms are efficient enough to run in real-time and at the same time delivery useful data to the user. As a part of this thesis, software for underwater inspections will be develop for the Australian Custom Service. This will in the most parts give extra good occasions where filters and algorithms can be tested in real situations and/or on material from real inspections.

1.  引言

2.  设计方法

3.  已有的工作

4.  分析

5.  结果

6.  参考文献

更多精彩文章请关注公众号：