摘要海洋面积占地球总面积的的百分之七十一，是支持地球上所有生命系统不可或 缺的一部分，在海洋中，藏着极其富有的天然矿物资源。近几十年来，由于陆地资 源的过度开采，人们开始把开采目标转移到海洋，世界各国对海洋的探索与开发就 日趋激烈，船舶行业的发展前景一片光明，无论何种作业方式，都离不开船舶。当 船舶在水中航行时，船底很容易与水中的绳索、垃圾等悬浮物接触，这些悬浮着的 物体便会缠绕在螺旋桨里，为安全航行制造了危险因素。因此，水下作业机器人的 研究与水下作业的检测无可厚非，水下声成像技术也越来越重要，这不仅减少了人 工作业中人力和物力的损耗，而且使水下探索和操作更加安全。84072

本文根据当今世界对水下机器人（ROV）的研究现状，进一步深入阐述水下检测 与作业机器人的原理，着重介绍各种声纳的工作原理以及声成像技术。以 852 型声 纳系统为例，对数据进行采集，通过一系列算法及公式变换生成原始图像。再对原 始图像进行线性灰度变换，为了突出后续处理工作中滤波的使用效果，先对灰度图 像添加椒盐噪声，然后再采用均值滤波方式进行滤波，对比均值滤波前后图像的清 晰度，最后采用阈值分割及边缘检测方法，对图像进行做进一步深入处理，使图像 特征更加容易被捕获。本次论文中所有的设计与仿真用的都是 MATLAB 软件，经过 多次设计与调试，最终得到了处理完的声纳图像，达到预期效果。

毕业论文关键词：水下机器人；声纳成像；图像处理

Abstract The ocean area accounts for seventy percent of the whole earth,which is an integral part of life on earth。In the sea,there is a wealth of natural mineral resources。In recent decades,due to the transition of the resources on the land,people began to transfer the mining target to the ocean。Countries around the world on the world on the exploration and development of the sea is increasingly fierce,and the development of the ship industry has a bright future。No matter what kind of the job,all cannot be separated  from the   ship。When

ship navigates in water ,the bottom is very easily with water，garbage and other suspended

solids contact。These floating objects will be wrapped around the propeller,and which will be a danger for safety navigation。Therefore,detection of underwater robot research and underwater operation is understandable,and underwater imaging technology is becoming more and more important。This technolo gy not only reduced the loss of manpower and material resources in manual operation,and underwater exploration and it is safer to operate。

In this paper,according to the research of underwater robot in current world,we will further elaborate the principle of underwater detection and operation of the robot,focuses on the working principle of a variety of sonar and acoustic imaging technology。Taking the 852 type sonar system as an example,the data is collected,and the original image is generated through a series of algorithms and formulas。Then we  can make original image do gray-scale transformation。 In order to highlight the subsequent processing of  the filtering effect,firstly we add ‘salt and pepper’ noise to the gray image。Then the mean filter, filtering and comparison before and after the clarity of the image。Finally, the maximum between class variance threshold segmentation method and edge detection method, for image enhancement features and more likely to be captured。In this paper,all of the design and    simulation   are    using   MATLAB   software。After   many   times   of   design   and

II

debugging,finally we get the sonar image processing,which has achieved the  desired results。