一种声图直航段数据自动提取及镶嵌的方法

A straight track sonar images auto-extracting and mosaicking method

声图镶嵌是利用成像声纳获得声图进行测深、水下地形地貌测绘的重要应用方向。常见的成像声纳多要求声纳载体做直线运动,而实际应用中载体有大量转弯的情况。为此提出了一种声图直航段数据自动提取方法,该方法首先利用Douglas-Peucker算法对声纳航迹进行抽稀处理,之后利用相邻抽稀点之间的距离判断数据是否为直航段,最后将直航段图像提取出来。获得直航段数据后,利用地理信息对声图进行镶嵌处理。对千岛湖湖底地貌图像进行了处理,获得了较好的镶嵌结果。

Sonar image mosaicking is an essential application in the underwater oceanographic survey. Usually imaging sonars assume that the platform traverses a perfectly straight track. However this is the seldom case since the sonar platforms often swerve for planned or unplanned reasons. Thus a method automatically extracting the sonar image segments whose tracks are straight is proposed. For a sonar track consisting of all the sonar positioning points, the Douglas-Peucker algorithm is applied to get a much smaller subset of points that could define the sonar track approximately. Then if the distance between two adjacent points in the subset is longer than a threshold, the relevant image is extracted as a straight segment. Finally the straight segments are mosaicked on an earth model based on their geographic positions. For validation the method is applied to segment and mosaic the real sonar images in the Qiandao Lake trial. Results show the affectivity of the proposed method.