Research On The Monitoring Technology For Sliding Launch Trajectory Of The Jacket

Since the dawning of21st century, with the exploration and exploitation of theocean, ocean engineering has been attracting more and more people’s attention, whichpromotes the rapid development of the detection and mining the oil stored deeplyunder the ocean. Meanwhile, the forecasting and monitoring of the influence underrelevant forces and subsequent reactions of the offshore oil platform and marinestructures subjected to the complicated ocean environment have been more focused onever. Jacket, which is a critical tool for exploiting the ocean, however, evolves to bemore awkward as extended to deeper ocean, simultaneously accompanied by thesoaring amount of money and human power required. In this sense, whether the jacketcan be laid into the water safely would directly affect the exploitation of the marineresources for our country; thereby researches on the monitoring technology for slidinglaunch trajectory of the jacket is particularly momentum to us.In this thesis, a method based on the theory of computer vision is utilized by thisresearch. First of all, obtaining the video information of the jacket’s sliding launchtrajectory by recording two cameras in order to obtain image sequences. Then usingthe basic theory of the stereoscopic vision to manipulate the image sequences, ofwhich the key points of the image can be mapped into the3d space by an improvedcamera calibration technique and Harris algorithm. Finally the six degrees of freedomcomponent values and motion attitude of the jacket in each time can be obtained. Weused a car as the model to do the experiment, and then compared the experimentalresults with the theoretical value. We found out that the experimental result isreasonable within acceptable error, which has proved that the method can be used tomonitor the rigid body trajectory. The study mainly includes:1. Introduction to the fundamental theory and basic knowledge of computervision, study on the camera model, principles and the calibration technology, andoptimizing the technology of the extraction match for feature points.2. Mathematical analysis and theoretical derivation of the technology of3dreconstruction and the method of six degrees of freedom decomposition; introducedMultiple View Geometry for the follow-up work preparation. 3. Used the method to do the car model experiment, compared the experimentalresult with the theoretical value, and then deduced conclusion.