Image Registration for Volume Measurement in 3D Range Data: For LHD Excavator Buckets in Underground Mining
Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
In this thesis algorithms have been designed based on morphological operators and the Harris method for corner detection to measure the volume of an region in a 3D image. This application is in the mining industry and enables the measurement of the volume of rocks carried in the bucket of an excavator based on a 3D range image of an empty and a full bucket. This work identifies the vehicle using morphological operators to delimitate the bucket followed by corner detection using using the Harris method combined with some additional modifications in order to improve its accuracy. Finally a 2D geometric transform is applied to match an image of a full bucket with an empty bucket to extract the rocks and measure its volume. Evaluation on 30 different bucket images was performed including some special cases resulting in a sufficiently fast time of execution of approximately 1 minute and a final error always below one cubic meter.
Place, publisher, year, edition, pages
2011. , 64 p.
Technology, image matching, Geometric transformation, 3D range data, volume measurement, mining
IdentifiersURN: urn:nbn:se:ltu:diva-54308Local ID: b4257191-9c54-4e60-94de-5024c8a2a5bcOAI: oai:DiVA.org:ltu-54308DiVA: diva2:1027689
Subject / course
Student thesis, at least 30 credits
Computer Science and Engineering, master's level
Validerat; 20110824 (anonymous)2016-10-042016-10-04Bibliographically approved