基于阈值分割的气幕围油栏气幕油液边缘识别  

Edge Identification of Air Curtain and Oil in Pneumatic Oil Barriers Based on Threshold Segmentation

作　　者：高启迪 卢金树 张高纶 GAO Qi-di;LU Jin-shu;ZHANG Gao-lun(School of Naval Architecture and Maritime,Zhejiang Ocean University,Zhoushan 316000,China)

机构地区：[1]浙江海洋大学船舶与海运学院,浙江舟山316000

出　　处：《电脑与信息技术》2022年第4期18-21,共4页Computer and Information Technology

基　　金：舟山市科技局项目(项目编号:2018C21008)。

摘　　要：近年来海上石油开发、运输等行业发展迅猛,给经济增长带来持续活力的同时,也使海上溢油事故频繁发生。气幕式围油栏作为一种新式的围油方法,成为当前防止溢油灾害的新兴热门手段。气幕式围油栏通过无人机采集溢油区域海上实时图像,进行图像处理,自动判断围油效果是否达到要求并进一步调控围油装置。为实现精准调控,需要精确且快速的图像处理算法即气幕、油液边缘识别与最小距离判定。文章通过阈值分割的方法,将原始图像与背景图像进行灰度阈值分割识别目标边缘信息,通过滤波降噪去除图像中的干扰像素,最终实现气幕与油液边缘的精准识别与两者最短距离的判定。In recent years, offshore oil development, transportation and other industries have developed rapidly, which brings sustained vitality to economic growth, but also makes offshore oil spill accidents occur frequently. As a new type of oil fence,pneumatic oil barriers have become a popular means to prevent oil spill disasters. The pneumatic oil barriers collect real-time images of the oil spill area at sea through UAV(Unmanned Aerial Vehicle), and conducts image processing to automatically determine whether the oil fence effect meets the requirements and further regulate the oil fence device. In order to achieve precise control, accurate and fast image processing algorithms are needed, namely, gas curtain and oil edge recognition and minimum distance determination. In this paper, through the method of threshold segmentation, the original image and the background image are segmented by gray threshold to identify the edge information of the target, and the interference pixels in the image are removed by filtering noise reduction. Finally, the accurate identification of the gas curtain and the oil edge and the determination of the shortest distance between the two are realized.

关 键 词：气幕式围油栏 图像处理 边缘识别 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]