Finding trees in the lidar point cloud

individual trees extracted from the lidar point cloudLi, W., Q. Guo, M. Jakubowski and M. Kelly. 2012. A new method for segmenting individual trees from the lidar point cloud. Photogrammetric Engineering and Remote Sensing 78(1): 75-84

In this study we develop a new algorithm to segment individual trees from the small footprint discrete return airborne lidar point cloud. The new algorithm adopts a top-to-bottom region growing approach that segments trees individually and sequentially from the tallest to the shortest. We experimentally applied the new algorithm to segment trees in a mixed coniferous forest in the Sierra Nevada Mountains in California, USA. Our results indicate that the proposed algorithm has good potential in segmenting individual trees in mixed conifer stands of similar structure using small footprint, discrete return lidar data.

Click to read more ...

Including the spatial relations of objects in classification

Liu, et al. 2008. ISPRS Journal of Photogrammetry and Remote Sensing. We establish a context where objects are areal (not points or lines) and non-overlapping (we call this “single-valued” space), and propose a framework of binary spatial relations between segmented objects to aid in object classification.

Click to read more ...

Spatial‐temporal tree mortality patterns associated with SOD

Liu, et al. 2007. Forest Ecology and Management. Inhomogeneous cross-K-functions of SOD and California bay trees: the peak in the dark line indicates co-clusteringWe analyzed the spatial–temporal patterns of overstory oak tree mortality in China Camp State Park, CA over 4 years using the point patterns mapped from high spatial resolution remotely sensed imagery. Both univariate and multivariate spatial point pattern analyses were performed (inhomogeneous K-functions and Neyman–Scott point processes) to characterize the spatial dependence among dead oak trees in each year, and between dead oaks and CA bay trees.

Click to read more ...

Modeling SOD risk in the United States

Kelly, et al. 2007. Computers, Environment and Urban Systems. Risk for SOD in the southeast based on model agreementUsing the locations of P. ramorum in CA we derived a risk map for SOD spread in the conterminous US using 5 environmental niche models: Expert-driven Rule-based, Logistic Regression, Classification and Regression Trees, Genetic Algorithms, and Support Vector Machines.

Click to read more ...

Obia for mapping dead trees

Guo, Q. C., M. Kelly, P. Gong and D. Liu. 2007. GIScience and Remote Sensing. An object, defined by spectral similarity of neighboring pixelsWe developed an object-based approach, including an image segmentation process and a knowledge-based classifier, to detect individual tree mortality in imagery of 1 m spatial resolution.

Click to read more ...

Automatic registration of airborne images

Liu, et al. 2006. Photogrammetric Engineering and Remote Sensing. Deformation vector plots overlaid on areas with high and low texture.Accurate registration of airborne images is challenging because complex local geometric distortions are often involved in image acquisition. We propose a solution to this registration problem in two parts: 1) an area-based method to extract sufficient numbers of well-located control points, 2) we use the extracted control points with local transformation models to register multi-temporal airborne images.

Click to read more ...

Support vector machines for predicting distribution of Sudden Oak Death in California

Guo, et al. 2005. Ecological Modeling. Predicted area of SOD risk in northern CaliforniaWe present an alternative method to conventional environmental niche modeling approaches by developing support vector machines (SVMs), which are the new generation of machine learning algorithms used to find optimal separability between classes within datasets, to predict the potential distribution of Sudden Oak Death in California.

Click to read more ...

Mapping oak mortality using high res CIR imagery

Kelly, et al. 2004. Photogrammetric Engineering and Remote Sensing. Supervised, unsupervised, and “hybrid” classification methods were evaluated for their accuracy in discriminating dead and dying tree crowns from bare areas and the surrounding forest mosaic utilizing 1-m ADAR imagery covering both tanoak/redwood forest and mixed hardwood stands. In both study areas the hybrid classifier significantly outperformed the other methods.

Click to read more ...

Interpretation of scale in paired quadrat variance methods

Guo and Kelly. 2004. Journal of Vegetation Science. Previous interpretations of the variance plot of paired quadrat variance method (PQV) have been incomplete; and in this study was to clarify the interpretation of PQV, and to shed additional light on how different quadrat variance methods can be used, in concert, to measure scale in transect data.

Click to read more ...