TY - GEN
T1 - Assessing woody vegetation cover dynamics in the Kruger National Park, South Africa
T2 - IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015
AU - Munyati, Christopher
AU - Sinthumule, Ndidzulafhi I.
N1 - Publisher Copyright:
© 2015 IEEE.
PY - 2015/11/10
Y1 - 2015/11/10
N2 - The coexistence of woody vegetation and grass characterises savannas. High spatial resolution digital satellite images and appropriate image processing algorithms can facilitate monitoring of the woody cover. Historical aerial photographs enable an extension of high spatial resolution analysis of the woody cover to periods before satellite imagery. This study tested a methodology that used historical (1998, 2012) SPOT-4 HRVIR gray scale ('red') band (0.61-0.68um, 10m resolution) and SPOT-5 HRG panchromatic band (0.48-0.71 urn, 5m resolution) images in combination with panchromatic (0.4-0.7um) aerial photographs (1940, 1968, 1977; 0.44-1.35m resolution) in monitoring woody cover at sites in the Kruger National Park, South Africa. Mean Euclidean Distance texture analysis in 3×3 moving windows enhanced woody cover. The woody cover on the respective multitemporal texture images was extracted through K-means clustering, with high (>85%) indicative classification accuracy. Boolean GIS overlay analysis of the resulting woody cover thematic layers enabled woody cover change detection.
AB - The coexistence of woody vegetation and grass characterises savannas. High spatial resolution digital satellite images and appropriate image processing algorithms can facilitate monitoring of the woody cover. Historical aerial photographs enable an extension of high spatial resolution analysis of the woody cover to periods before satellite imagery. This study tested a methodology that used historical (1998, 2012) SPOT-4 HRVIR gray scale ('red') band (0.61-0.68um, 10m resolution) and SPOT-5 HRG panchromatic band (0.48-0.71 urn, 5m resolution) images in combination with panchromatic (0.4-0.7um) aerial photographs (1940, 1968, 1977; 0.44-1.35m resolution) in monitoring woody cover at sites in the Kruger National Park, South Africa. Mean Euclidean Distance texture analysis in 3×3 moving windows enhanced woody cover. The woody cover on the respective multitemporal texture images was extracted through K-means clustering, with high (>85%) indicative classification accuracy. Boolean GIS overlay analysis of the resulting woody cover thematic layers enabled woody cover change detection.
KW - Change detection
KW - Geospatial analysis
KW - Image classification
KW - texture analysis
KW - vegetation mapping
UR - http://www.scopus.com/inward/record.url?scp=84962519493&partnerID=8YFLogxK
U2 - 10.1109/IGARSS.2015.7326762
DO - 10.1109/IGARSS.2015.7326762
M3 - Conference contribution
AN - SCOPUS:84962519493
T3 - International Geoscience and Remote Sensing Symposium (IGARSS)
SP - 4240
EP - 4243
BT - 2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 26 July 2015 through 31 July 2015
ER -
