Automate green coverage measure using a novel DIA method: UIP-MGMEP
Green coverage is an important indicator of healthstatus of rangeland, and its meaningful to improve the efficiency,accuracy and objectivity of green coverage measurement. Digitalimage analysis (DIA) technique becomes more attractive;however, there have been few advances in automaticmeasurement methods. In this paper, we automate rangelandgreen coverage measure by DIA, which is a novel thresholdsegmentation algorithm named as UIP-MGMEP applied to thevegetation index (Excess Green (EXG) in this paper).UIP-MGMEP is not recommended to images whose fraction ofvegetation is smaller than 0.5% or larger than 99.5%,which could be rounded to 0% or 100% visually.UIP-MGMEP optimizes EXG threshold by searching the upperinflexion point (UIP) of the M-Et curve (mean gradientmagnitude of edge pixels (MGMEP) vs EXG threshold), based onthe assumption that EXG variance of either background orvegetation is suppressed. The results show that UIP-MGMEPworks well, and fails when either background or vegetation is toocomplex. UIP-MGMEP achieves accurate and objective greencoverage measure without any human intervention. However,UIP-MGMEP is only developed to extract green-leavedvegetation, not suitable for non-green (even grayish-green) leavedvegetation. Accuracy could be improved through using VIS-NIRcamera instead of VIS camera.
Jianbo Hu Wei Huang Tao Lu Wei Chen Xingyuan He
Laboratory of Environmental Protection in Water Transport Engineering, Tianjin Research Institute fo Xinxian development and reform commission, Xinyang, China Institute of Applied Ecology, Chinese Academy of Sciences, Shenyang, China
国际会议
The 19th International Conference on Geoinformatics(第19届国际地理信息科学与技术大会 19 Geoinformatics)
上海
英文
1-5
2011-06-01（万方平台首次上网日期，不代表论文的发表时间）