Project description
Nature of project: experimental, data analysis
Available to full-time physicists .
Best suited to students on these degree schemes:
Wave energy will become an especially important area of renewable resources in the next few years. However in order to be able to site wave energy projects, a cheap and effective method must be found to monitor wave heights.Your task will be to use video images of buoys in Cardigan bay in order to measure wave heights. The video imagery will be acquired through a robotic telescope, whose position and orientation (dip angle) with respect to the sea surface will be known. Using some simple trigonometry, the distance to the buoy, and image scale (cm/pixel) can be determined. The top or centre of figure of the buoy will be used as a reference point to measure wave heights.Unfortunately two effects will limit the accuracy of measuremnts. Firstly wind shake on the telescope tube, and secondly turbulence in the atmosphere. The former can be compensated for by placing a landmark on the shoreline into the same field of view as the bouy. The latter (and possibly the former too) could possibly be removed by applying an Fourier transform to the wave height time sequence data in order to be able to separate low frequency wave heights from the higher frequency atmospheric turbulence. You will also have to model tidal height and possibly also refractive effects if observing buoys close to the horizon. Can you use opportunistic targets instead of bouys such as flotsam, floating sea gulls, or jelly fish? How will you estimate accuracies of measurement? [edit]
Currently with the telescope, once it is pointing away from the shore line, we lose the ability to identify where on the sea we are looking and the telescope orientation encoders are not suffiently accurate to be relied upon to give us a precise azimuth and elevation angle.On top of the telescope tube is a separate wide angle finder scope camera, which will cover the area of the sea being imaged by the telescope, but also fixed landmarks of know position on the shore line.If you can calibrate/calculate a rotation and scaling matrix to go from telescope camera pixel coordinates into finder scope camera pixel coordinates (and vice versa), then anything that you find in the sea, through the telescope, you can find its precise location in the finder scope. If the finderscope contains landmarks on the shore line of known longitude, latititude and height, then you should be able to calculate a precise bearing and range to the objsect in the sea - hence solving our location problem.Another aspect of the project that needs inproving is the ability to track objects in the sea automatically, This would be done by finding the centre of figure of the object over time, then applying a slewing movement to the telescope, using ASCOM software, to recentre the object in the image again. Some good programing skills would be needed for the image analysis part of this. [edit]
The project difficulty will depend upon how far you wish to work away from the shore. Close to the shoire you have landmarks that can be used for removing wind shake. However the further out that you go towards the horizon, the more that you must consider the effects of terrestrial refraction on the dip angle and the effects of tidal heights on the image scale.Some programming may be neccessary in IDL or LabView in order to track and measure buoy (or other target) heights. If you do not do this then manual measurments will be acceptable, although somewhat manually intensive to make.Help will be given on telescope training and an introduction to video data capture and processing. [edit]
A robotic telescope on top of the IMAPS building.A lab PC. [edit]
| milestone | to be completed by |
|---|---|
| Telescope training | end of October |
| Preliminary data capture | Christmas |
| Camera modelling and wave height measurements | end of February |
| Lengthy trials on different targets in order to determine accuracy of measurement | Easter |
Blame me, not my employer etc. etc.