基于光学遥感图像的目标检测技术研究-信号与信息处理专业论文.docxVIP

Abstract一一一_————————————————————————————_————————、————————一 Abstract 一一一_————————————————————————————_————————、————————一 ABSTRACT As human beings paid more and more attention to the sea，ship became the crucial tool to utilize and explore the sea．In order to get more information of ship， the ship detection technology was developed and applied to fields such as monitoring traffic，fishery,succor in the sea，anti—terrorism and anti—pirate． Compared with the traditional ship detection based on SAR，the ship detection based on the optical imams was more difficult because of the imaging system of the optical sensors，while the related papers ignored．So，it was very meaningful to study ship detection algorithm based on the optical images． The impact of the problems above on the false alarm rate and missed alarm rate would be far-reaching．Aiming at those problems，we designed the srdp detection algorithm based on optical images through studying related papers，then built a prototype system based on GoogleEarth platform． Main contents and innovations of this dissertation were listed aS following： 1．W．e proposed a new segmentation algorithm of sea and land based on 3D reconstruction(ssL一3DR)．Compared with the traditional 2D algorithm,our algorithm portrayed the land area better and was more robust． 2．Aiming at the cloud detection，we proposed a coment based image anysis-feedback learning algorithm for cloud detection(CBIA-FLCD)，which improved the accuracy and the robustness of cloud detection． 3．We finally proposed a new ship detection approach for optical remote sensing images via multilevel vision perception—dissymmetrical allocation of visual attention rSDMVP．DAVA)．Experiment results showed that our method can more effectively detect ship targets compared the traditional methods，especially for the complex sea surface background． 4．We designed the ship detection system based on the Google Earth platform realizing algorithms above，which had hiper detection rate compared with the tra