Change search
ReferencesLink to record
Permanent link

Direct link
Collision detection and avoidance system based on computer vision
Norwegian University of Science and Technology, Faculty of Information Technology, Mathematics and Electrical Engineering, Department of Engineering Cybernetics.
2014 (English)MasteroppgaveStudent thesis
Abstract [en]

A wide selection of stereo matching algorithms have been evaluated for the purpose of creating a collision avoidance module. Varying greatly in the accuracy, a few of the algorithms were fast enough for further use. Two computer vision libraries, OpenCV and MRF, were evaluated for their implementations of various stereo matching algorithms. In addition OpenCV provides a wide variety of functions for creating sophisticated computer vision programs and were evaluated on this basis as well. A stereo camera were constructed using low cost, of-the-shelf web cameras. Two low-power platforms, The Pandaboard and the Beaglebone Black, were evaluated as viable platforms for developing a computer vision module on top. In addition they were compared to an Intel platform as a reference. Based on the results gathered, a fast, but simple, collision detector could be made using the simple block matching algorithm found in OpenCV. A more advanced detector could be built using semi-global stereo matching. These were the only implementations that were fast enough. The other energy minimization algorithms (Graph cuts and belief propagation) did produce good disparity maps, but were too slow for any realistic collision detector. In order for the low-power platforms to be fast enough, a combination of improvements must be used. OpenCV should be compiled with aggressive optimization options enabled with support for hardware accelerated floating point math. Choice of low-power platform matters, but it is possible to work around this by reducing the workload. The most effective speedup that enables the low-power platforms were reducing the resolution of the images to be matched. When reducing the size of the sub-problems enough to align with cache size, considerable speedups were found with little penalty in the corresponding disparity map.

Place, publisher, year, edition, pages
Institutt for teknisk kybernetikk , 2014. , 107 p.
URN: urn:nbn:no:ntnu:diva-24779Local ID: ntnudaim:7123OAI: diva2:720034
Available from: 2014-05-27 Created: 2014-05-27 Last updated: 2014-05-27Bibliographically approved

Open Access in DiVA

fulltext(9358 kB)544 downloads
File information
File name FULLTEXT01.pdfFile size 9358 kBChecksum SHA-512
Type fulltextMimetype application/pdf
cover(3257 kB)5 downloads
File information
File name COVER01.pdfFile size 3257 kBChecksum SHA-512
Type coverMimetype application/pdf
attachment(139639 kB)30 downloads
File information
File name ATTACHMENT01.zipFile size 139639 kBChecksum SHA-512
Type attachmentMimetype application/zip

By organisation
Department of Engineering Cybernetics

Search outside of DiVA

GoogleGoogle Scholar
Total: 544 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: 470 hits
ReferencesLink to record
Permanent link

Direct link