TY - GEN
T1 - Correction of building height effect using LIDAR and GPS
AU - Sohn, Hong Gyoo
AU - Yun, Kong Hyun
AU - Kim, Gi Hong
AU - Park, Hyo Sun
PY - 2005
Y1 - 2005
N2 - Correction of building height effects is a critical step in image interpretation from aerial imagery in urban area. In this paper, an efficient scheme to correct building height effects from aerial color imagery using multi source data sets is presented. The following steps have been performed to remove the shadow effect from aerial color imagery. First, the shadow regions of the orthorectified aerial color image are precisely located using the solar position and the heights of ground objects derived from LIDAR (Light Detection and Ranging) data. The shadow area is composed of many different ground features. To accurately recover the original spectral information the step for accurate segmentation of shadow regions needed. This step has been performed by utilizing the existing digital map which contains surface cover information. To assign correction information corresponding to the surface features on the shadowed area, correction factor needed to be modeled. For this three basic assumptions are proposed and comparison between the context region and the same non-shadowed context region is made. Finally, the shadow-effect-corrected image is generated by using the correction factor obtained from the previous step.
AB - Correction of building height effects is a critical step in image interpretation from aerial imagery in urban area. In this paper, an efficient scheme to correct building height effects from aerial color imagery using multi source data sets is presented. The following steps have been performed to remove the shadow effect from aerial color imagery. First, the shadow regions of the orthorectified aerial color image are precisely located using the solar position and the heights of ground objects derived from LIDAR (Light Detection and Ranging) data. The shadow area is composed of many different ground features. To accurately recover the original spectral information the step for accurate segmentation of shadow regions needed. This step has been performed by utilizing the existing digital map which contains surface cover information. To assign correction information corresponding to the surface features on the shadowed area, correction factor needed to be modeled. For this three basic assumptions are proposed and comparison between the context region and the same non-shadowed context region is made. Finally, the shadow-effect-corrected image is generated by using the correction factor obtained from the previous step.
UR - http://www.scopus.com/inward/record.url?scp=33646422989&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=33646422989&partnerID=8YFLogxK
U2 - 10.1007/11557654_121
DO - 10.1007/11557654_121
M3 - Conference contribution
AN - SCOPUS:33646422989
SN - 3540290311
SN - 9783540290315
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 1087
EP - 1095
BT - High Performance Computing and Communcations - First International Conference, HPCC 2005, Proceedings
T2 - 1st International Conference on High Performance Computing and Communcations, HPCC 2005
Y2 - 21 September 2005 through 23 September 2005
ER -