基于机器视觉的UUV水下沉底目标自主识别
UUV Autonomous Recognition of Underwater Bottom Target Based on Machine Vision
庞彦东 1曹子健 2文无敌 1张伽伟 1姬庆 1张森 3李太伟 1张志强1
作者信息
- 1. 海军工程大学兵器工程学院 武汉 430070
- 2. 中国人民解放军91837部队 舟山 316000
- 3. 海军工程大学电子工程学院 武汉 430070
- 折叠
摘要
水下无人航行器对目标的自动识别研究是未来水下目标探测领域的发展趋势.针对现阶段水下沉底目标声呐图像探测人工识别工作量大、判别困难等问题,研究了机器视觉中基于Canny算子图像边缘检测结合霍夫线变换去噪的识别方法在水下无人航行器自主探测的应用,为此在湖底布放多个水下沉底模拟目标,尝试了水下无人航行器的目标自主探测方案,通过实际湖上试验对比分析三种水下沉底目标声呐图像的算法识别效果,验证了论文所研究方法的有效性,为后续无人平台自主识别算法研究提供参考,以期提高无人航行器对水下沉底目标的自主探测识别能力.
Abstract
Automatic target recognition of underwater unmanned vehicles(UUV)is a developing trend in the field of underwa-ter target detection in the future.In view of the large amount of manual identification work and difficulty in discriminating the under-water sonar image detection at the present stage,the application of the recognition method based on Canny operator image edge de-tection combined with Hough line transform denoising in the autonomous detection of underwater unmanned vehicles in machine vi-sion is studied.For this purpose,a number of underwater simulation targets are deployed in Lake.This paper attempts the autono-mous target detection scheme of underwater unmanned vehicles,and compares and analyzes the algorithm recognition effect of three kinds of underwater sonar images through the actual lake test,which verifies the effectiveness of the proposed method and provides reference for the subsequent research on autonomous recognition algorithm of unmanned platforms,in order to improve the autono-mous detection and recognition ability of underwater targets of unmanned vehicles.
关键词
水下无人航行器/机器视觉/数学形态学/声呐图像Key words
underwater unmanned vehicles/machine vision/mathematical morphology/sonar image引用本文复制引用
出版年
2024
国家科技期刊平台
NSTL
万方数据