针对海底电缆、水下管道等水下线状目标的自动光电探测需求，基于距离选通成像探测方法，构造实验系统在实验船池环境下对水下线状目标进行探测实验，根据图像特征，采用了结合中值滤波、对比度拉伸和小波去噪的图像预处理方法改善图像质量，采用改进型Canny边缘检测算法识别线状目标边界，采用改进型RANSAC算法对水下线状目标进行曲线拟合。实验结果表明本文提出的方法可在5倍水下衰减长度内，有效用于二次曲线状目标检测，算法稳定可靠，可适用于任意形状的水下线状目标光电探测应用。
In order to meet the demand of underwater linear objects detection, such as underwater cable and pipeline, an experimental device based on underwater range gated imaging method was built in this article. Some experiments under experimental pool conditions were carried out with this device. According to the characteristics of the result images, an algorithm which combined median filter, contrast stretching and wavelet denoising method was employed to improve the image quality. In order to detect the underwater linear object's edge, the improved Canny edge detection algorithm was employed. The improved RANSAC algorithm was employed for underwater target linear curve fitting. The experimental results show that the proposed method can be within the attenuation length of 5 times the underwater, effectively used for target detection, the algorithm is stable and reliable, and can be applied to any shape of the underwater target linear photoelectric detection applications
2018,40(10): 130-134 收稿日期：2018-03-31
DOI：10.3404/j.issn.1672-7649.2018.10.026
分类号：TN249
作者简介：姜朝宇(1979-),男,工程师,主要从事通信、声呐技术研究
参考文献：
[1] ANTICH J, ORTIZ A. Underwater cable tracking by visual feedback[C]//Iberian Conference on Pattern Recognition and Image Analysis. Berlin:Springer Press, 2003. 53-61.
[2] NGUYEN H G, HECKMAN P J, PAI A L. Real-time pattern recognition for guidance of an autonomous undersea submersible[C]//IEEE International Conference on Robotics and Automation. New York:IEEE Press, 1988. 1767-1770.
[3] HALLSET J O. Simple vision tracking of pipelines for an autonomous underwater vehicle[C]//IEEE International Conference on Robotics and Automation, New York:IEEE Press, 1991:2767-2772.
[4] CHEN H H, CHUANG W N, WANG C C. Vision-based line detection for underwater inspection of breakwater construction using an ROV[J]. Ocean Engineering, 2015, 109(15):20-33.
[5] 朱炜, 徐玉如, 万磊, 等. 基于二维直方图和粒子群优化的边缘检测[J]. 系统工程与电子技术, 2007, 29(7):1192-1196.
[6] 朱炜, 徐玉如, 秦再白. 基于PSO和模糊划分熵的水下图像分割[J]. 光学技术, 2007, 33(5):754-758.
[7] GONZALEZ C, WOODS E, Digital image processing[M]. New Jersey:Prentice Hall, 2007.
[8] 徐涛, 杨克成, 夏珉, 等. 基于距离选通成像的水下线状目标检测算法[J]. 激光与红外, 2017, 47(10):1321-1324.
