dc.contributor.author |
Cronje, J
|
|
dc.contributor.author |
De Villiers, J
|
|
dc.date.accessioned |
2013-01-31T06:53:22Z |
|
dc.date.available |
2013-01-31T06:53:22Z |
|
dc.date.issued |
2012-11 |
|
dc.identifier.citation |
Cronje, J and De Villiers, J. 2012. A comparison of image features for registering LWIR and visual images. 23rd Annual Symposium of the Pattern Recognition Association of South Africa (PRASA), Pretoria, South Africa, 29-30 November 2012 |
en_US |
dc.identifier.isbn |
978-0-620-54601-0 |
|
dc.identifier.uri |
http://www.prasa.org/proceedings/2012/prasa2012-37.pdf
|
|
dc.identifier.uri |
http://hdl.handle.net/10204/6502
|
|
dc.description |
23rd Annual Symposium of the Pattern Recognition Association of South Africa (PRASA), Pretoria, South Africa, 29-30 November 2012 |
en_US |
dc.description.abstract |
This paper presents a comparison of several established and recent image feature-descriptors to register long wave infra-red images in the 8–14 m band to visual band images. The feature descriptors were chosen to include robust algorithms, SURF and SIFT — and fast algorithms, BRISK and BFROST. To evaluate the feature-descriptors a ground truth was created by determining the intrinsic and extrinsic camera calibration parameters for the cameras and using this to photogrammetrically relate pixel positions between the images. The inlier results of each feature descriptor for the top 20%, 50% and 100% of the matches (based on match strength) were used to create a homography. The average pixel error between the homography reprojected feature points and the photogrammetric reprojection was used as the error. The results show that none of the descriptors perform well in standard form, with BFROST faring slightly better than the other algorithms. This suggests a need to modify the algorithms to detect physical/structural features and de-emphasise textural features. |
en_US |
dc.language.iso |
en |
en_US |
dc.publisher |
PRASA |
en_US |
dc.relation.ispartofseries |
Workflow;9973 |
|
dc.subject |
Long Wave Infra Red |
en_US |
dc.subject |
LWIR |
en_US |
dc.subject |
LWIR images |
en_US |
dc.subject |
Algorithms |
en_US |
dc.subject |
Visual images |
en_US |
dc.title |
A comparison of image features for registering LWIR and visual images |
en_US |
dc.type |
Conference Presentation |
en_US |
dc.identifier.apacitation |
Cronje, J., & De Villiers, J. (2012). A comparison of image features for registering LWIR and visual images. PRASA. http://hdl.handle.net/10204/6502 |
en_ZA |
dc.identifier.chicagocitation |
Cronje, J, and J De Villiers. "A comparison of image features for registering LWIR and visual images." (2012): http://hdl.handle.net/10204/6502 |
en_ZA |
dc.identifier.vancouvercitation |
Cronje J, De Villiers J, A comparison of image features for registering LWIR and visual images; PRASA; 2012. http://hdl.handle.net/10204/6502 . |
en_ZA |
dc.identifier.ris |
TY - Conference Presentation
AU - Cronje, J
AU - De Villiers, J
AB - This paper presents a comparison of several established and recent image feature-descriptors to register long wave infra-red images in the 8–14 m band to visual band images. The feature descriptors were chosen to include robust algorithms, SURF and SIFT — and fast algorithms, BRISK and BFROST. To evaluate the feature-descriptors a ground truth was created by determining the intrinsic and extrinsic camera calibration parameters for the cameras and using this to photogrammetrically relate pixel positions between the images. The inlier results of each feature descriptor for the top 20%, 50% and 100% of the matches (based on match strength) were used to create a homography. The average pixel error between the homography reprojected feature points and the photogrammetric reprojection was used as the error. The results show that none of the descriptors perform well in standard form, with BFROST faring slightly better than the other algorithms. This suggests a need to modify the algorithms to detect physical/structural features and de-emphasise textural features.
DA - 2012-11
DB - ResearchSpace
DP - CSIR
KW - Long Wave Infra Red
KW - LWIR
KW - LWIR images
KW - Algorithms
KW - Visual images
LK - https://researchspace.csir.co.za
PY - 2012
SM - 978-0-620-54601-0
T1 - A comparison of image features for registering LWIR and visual images
TI - A comparison of image features for registering LWIR and visual images
UR - http://hdl.handle.net/10204/6502
ER -
|
en_ZA |