针对无人潜航器搭载声呐在探测沉底小目标时存在混响背景强、目标边缘模糊和存在声影等特征，提出一种多尺度匹配的声呐图像识别方法。首先对声呐图像进行归一化处理，得到声呐回波强度的灰度图像；然后进行高斯低通滤波，增强了强回波区的轮廓特征；进一步采用自适应阈值法分割图像，形成相对独立的连通域；最后根据典型柱状目标在声呐图像中的尺寸、回波强度和声影强度等多尺度特征，提出目标多尺度匹配的声呐图像目标识别方法；将该方法用于不同声呐图像的目标识别，结果表明，利用多尺度匹配的声呐图像识别方法对典型沉底小目标具有较好的识别效果，为水下机器人目标自动识别提供一种有效方法。
A multi-scale matching sonar image recognition method is proposed to address the characteristics of reverberation background, blurred target edges, and the presence of sound shadows in unmanned underwater vehicles. Normalize the sonar image to obtain a grayscale image of the sonar echo intensity. Gaussian low-pass filtering was applied to enhance the contour features of the strong echo area. Adaptive threshold method was adapted to segment images and form relatively independent connected domains. Finally, a multi-scale matching model for target recognition in sonar images is proposed based on the multi-scale characteristics such as size, echo intensity, and sound shadow intensity of typical cylindrical targets in sonar images. The method was applied on different sonar images, and the results showed that the multi-scale matching sonar image recognition method has good recognition effect on typical small targets on the seabed. This providing an effective method for automatic target recognition of underwater robots.
2024,46(16): 120-124 收稿日期：2023-12-04
DOI：10.3404/j.issn.1672-7649.2024.16.019
分类号：TP242
作者简介：李荣（1976 – ），男，硕士，副教授，研究方向为水中兵器技术
参考文献：
[1] 张家铭, 丁迎迎. 基于深度学习的声呐图像目标识别[J]. 舰船科学技术, 2020, 42(12): 133-136.
ZHANG Jiaming, DING Yingying. Sonar image target recognition based on deep learning[J]. Ship Science and Technology, 2020, 42(12): 133-136.
[2] 何义才, 赵建虎, 张红梅, 等. 联合NSCT与多重分形的高噪声侧扫声呐图像分割[J]. 测绘学报, 2020, 49(2): 162-170.
HE Yicai, ZHAO Jianhu, ZHANG Hongmei, etal. Segmentation of polluted SSS image by combining NSCT and multifractal[J]. Acta Geodaetica et Cartographica Sinica, 2020, 49(2): 162-170.
[3] 韩婷婷, 王璐瑶, 周天, 等. FCM-CV水平集算法在沉底小目标声呐图像分割中的应用[J]. 水下无人系统学报, 2021, 29(3): 278-285.
HAN Tingting, WANG Luyao, ZHOU Tian, etal. Application of FCM-CV level set algorithm in sonar image segmentation of small sinking target[J]. Journal of Unmanned Undersea Systems, 2021, 29(3): 278-285.
[4] 刘琳, 李荣, 石剑. 基于伪彩色图像处理的猎雷声呐水雷目标检测技术[J]. 水雷战与舰船防护, 2016, 24(2): 28-31.
LIU Lin, LI Rong, SHI Jian. Mine detecting technique of mine-hunting sonar based on pseudo-color image processing[J]. Mine Warfare & Ship Self - Defence, 2016, 24(2): 28-31.
[5] 田原嫄. 基于小波包的声呐图像分割方法研究[D]. 吉林: 东北电力大学, 2021.
[6] 吴政峰, 张政, 袁明新, 等. 融合修正OTSU和中值滤波的水上航行器障碍物视觉分割[J]. 兵工自动化, 2020(7): 16-19.
WU Zhengfeng, ZHANG Zheng, YUAN Mingxin, et al. Visual segmentation incorporating modified otus and median filtering for obstacles of a watercraft[J]. Ordnance Industry Automation, 2020(7): 16-19.
[7] 苗锡奎, 朱枫, 许以军, 等. 基于视觉的水雷目标识别方法研究[J]. 海洋工程, 2012, 30(4): 154-160.
MIAO Xikui, ZHU Feng, XU Yijun, et al. Mine object recognition method research of AUV based on vision[J]. The Ocean Engineering, 2012, 30(4): 154-160.
[8] 盛蕴霞, 霍冠英, 刘静. 基于超像素聚类的侧扫声呐图像分割算法[J]. 计算机工程, 2018, 44(6): 225-231.
SHENG Yunxia, HUO Guanying, LIU Jing. Side-scan sonar image segmentation algorithm based on super-pixels clustering[J]. Computer Engineering, 2018, 44(6): 225-231.
