基于无人机遥感技术的港湾养殖区监测  

Monitoring of Inshore Aquaculture Cages Using Unmanned Aerial Vehicle

作　　者：闫静[1] 张彩云[1,2] 张永年 李雪丁 张宁[2] 江山[2] 商少凌[1,2] 

机构地区：[1]厦门大学水声通信与海洋信息技术教育部重点实验室,福建厦门361005 [2]厦门大学近海海洋环境科学国家重点实验室,福建厦门361102 [3]福建省测绘院,福建福州350003 [4]福建省海洋预报台,福建福州350003

出　　处：《厦门大学学报（自然科学版）》2016年第5期742-748,共7页Journal of Xiamen University：Natural Science

基　　金：国家科技支撑计划(2013BAB04B01);厦门大学海洋科学基地科研训练及科研能力提高项目(J1210050);厦门大学大学生创新创业训练计划(201510384091)

摘　　要：无人机(unmanned aerial vehicle,UAV)遥感具有高时效、机动灵活、高分辨率的特点,探索应用UAV遥感平台对沿海养殖区进行监测,实时掌握养殖发展动态,开展灾后应急监测,显然具有重要意义.利用UAV获取的CCD相机光谱信息,建立渔排监测指数,提出了渔排监测指数阈值快速提取算法;并针对渔排本身混合了水体和木筏2种目标物的特性,对提取结果进行图像闭合算子运算.在福建沿海古雷和可门港的实验结果表明,基于UAV获取的数据影像,利用渔排监测指数快速提取算法估算的渔排面积精度在两个区域分别达到98.9%和97.4%,与传统遥感分类方法——最大似然法相比,在同一计算平台上,计算速度提高5倍,无论精度或速度都可满足海洋与渔业管理部门在业务上的快速响应需求.此外应用UAV,以苏迪罗台风过后的三都澳海区为例,尝试开展了台风灾害造成的渔排损毁分析,初步推算出台风造成的渔排损毁面积比例.该研究表明利用UAV遥感可以为渔业管理提供有效的技术支持.Unmanned aerial vehicle（ UAV） may provide a quick means to gather information of inshore aquaculture cages for the government to have an emergency response to events such as typhoons.Based on CCD imageries with 0. 1 meter resolution acquired with UAV platform and its spectral features,here we proposed a threshold method to quickly detect the distribution of aquaculture cages in two bays of southern Fujian.An identification accuracy of 98. 9% and 97. 4%,for Gulei and Kemen bays,respectively,was obtained.Meanwhile,the image processing time is reduced by 5 times,compared to the time needed using the widely accepted conventional approach-maximum likelihood method.In addition,a rough evaluation on the ruin extent of aquaculture cages by typhoon was also attempted,using UAV acquired imageries and satellite images.This study suggests that optical remote sensing with UAV holds promise for better management of inshore aquaculture activities.

关 键 词：无人机(UAV) 渔排 快速检测 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]