Change search
ReferencesLink to record
Permanent link

Direct link
Continuous Autofocus for Line Scanning Hyperspectral Camera
Norwegian University of Science and Technology, Faculty of Information Technology, Mathematics and Electrical Engineering, Department of Electronics and Telecommunications.
2012 (English)MasteroppgaveStudent thesis
Abstract [en]

Good focus quality is essential in imaging applications. For a hyperspectral line scanning camera, lack of continuous autofocus will quickly make the image become out of focus, given curved or sloped objects. In spite of this, continuous autofocus solutions for hyperspectral line scanning cameras seem largely absent. In the presented work a continuous autofocus system for a HySpex VNIR-1600 hyperspectral line scanning camera is detailed. Passive autofocusing techniques were first tested, showing little potential due to failure of differentiating between contrast changes and focus quality. Moving on, an active autofocus solution was developed. Using a laser displacement sensor mounted ahead of the camera's field of view, the topography of the object is measured and focus adjusted accordingly. The system is dependent on several calibrations. Calibration procedures were invented to ease the calibration process, and the obtained calibration accuracies are discussed. Several samples were tested for focus quality using two different speed settings, and the quality and measured performance is discussed. Furthermore, working conditions for the laser sensor is also investigated. The resulting continuous autofocus system worked as intended, adjusting for topography changes within the physical limits of the system. The obtained focus quality for slow topography changes is excellent. For large topography changes over a small distance the system struggles to follow. Furthermore, adjustment for changes of 5mm over a distance of one refocusing period (4.1-4.3mm) is not recommended due to vibrations in the camera. The refocusing period for the system is off from the set value. While not impacting the functionality directly, it indicates an implementation error in the system. Implementing a dynamic refocusing period might improve system performance. Furthermore, "pause and adjust"-scanning is suggested as a possible improvement with potential for objects with sudden topography changes. Last, a more robust rig is advisable in order to prevent camera vibrations interfering with the image quality.

Place, publisher, year, edition, pages
Institutt for elektronikk og telekommunikasjon , 2012. , 114 p.
URN: urn:nbn:no:ntnu:diva-21101Local ID: ntnudaim:7258OAI: diva2:629225
Available from: 2013-06-16 Created: 2013-06-16 Last updated: 2013-06-22Bibliographically approved

Open Access in DiVA

fulltext(16011 kB)2735 downloads
File information
File name FULLTEXT01.pdfFile size 16011 kBChecksum SHA-512
Type fulltextMimetype application/pdf
cover(735 kB)17 downloads
File information
File name COVER01.pdfFile size 735 kBChecksum SHA-512
Type coverMimetype application/pdf
attachment(40368 kB)197 downloads
File information
File name ATTACHMENT01.zipFile size 40368 kBChecksum SHA-512
Type attachmentMimetype application/zip

By organisation
Department of Electronics and Telecommunications

Search outside of DiVA

GoogleGoogle Scholar
Total: 2735 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 558 hits
ReferencesLink to record
Permanent link

Direct link