科研之友
微信
新浪微博
Facebook
分享链接
摘要
The estimations of water column depth and the towfish altitude and the measurements of the target's location and shape all depend on the accuracy and reliability of bottom tracking of a sidescan sonar (SSS) waterfall image. Traditionally, the threshold method has often been adopted, but it is difficult to achieve an ideal bottom-tracking result due to contamination of the water column image in complicated measuring environments such as suspended solids and strong absorption sediments. Besides, it is also time consuming and easy to be influenced by operators' experience because the threshold values used need to be set segmentally. To track the bottom accurately and efficiently, first, this paper studies the threshold method, the last peak method, the assumption of seabed continuous variation, and the symmetry assumption. Then, a comprehensive bottom-tracking method is proposed. To validate the proposed method, an SSS experiment was conducted in the Bohai Sea of China. The bottom tracking for the SSS images was carried out automatically by the developed software. Comparing the port and starboard towfish altitudes tracked for all SSS images, the proposed method achieves +/- 0.19 m of the standard deviation referenced to water depth at the points of measurement. Furthermore, taking the integration of the sounding data, the tidal level, and the towfish depth as reference, the proposed method obtains +/- 0.17 m of the standard deviation. These parameters show that the proposed method has high accuracy in bottom tracking. Finally, this method is further discussed in the applications, and the results show that it has better performance than the traditional threshold method where the SSS waterfall image is influenced under complicated measurement environment. A shortcoming induced by the SSS imaging mechanism is also found that this method may be invalid when high seabed targets lie at two sides of towfish nadir, which can be solved by increasing the towfish altitude.
- 出版日期2017-7
- 单位武汉大学
