Authors: Ruiliang Pu*, UNIVERSITY OF South Florida
Topics: Remote Sensing, Biogeography, Applied Geography
Keywords: Spectral feature, Texture measure, Pleiades, Phenology, High resolution Satellite Imagery, Canonical correlation analysis.
Session Type: Paper
Presentation File: No File Uploaded
The forest LAI is an important structural parameter directly affecting terrestrial plant ecosystems. Existing studies on using very high resolution (VHR) multitemporal satellite imagery to investigate the seasonal effect on forest LAI at a landscape scale are rare. In this study, we proposed to map and analyze forest LAI using four seasonal Pléiades images over a natural forest area. A subset of selected spectral/textural features was used to develop pixel-based seasonal LAI regression models through a two-step feature selection procedure. Finally, LAI seasonal changes and mapping accuracies were analyzed and compared among the four season images. Experimental results indicate: (i) a set of optimal texture parameters for extracting the 1st- and 2nd-order gray level statistical textures was determined as a window size 5×5, a direction 900 and pixel displacement 4 pixels; (ii) textural features were more important than spectral features, and red band has a higher power in mapping forest LAI; (iii) the late spring Pleiades image resulted in the highest accuracy for estimating forest LAI; (iv) the maximum variation of forest LAI could be obtained with a typical summer season image; and (v) there exists a significant seasonal change of forest canopy LAI and the seasonal effect on forest LAI mapping can be assessed by using the multi-seasonal VHR satellite imagery at a landscape scale. This study is the first time using both spectral and textural information extracted from the multi-seasonal VHR images to evaluate the seasonal effect on forest canopy LAI mapping at a landscape scale.
To access contact information login