Object-Oriented Vegetation Classification at Peninsular National Parks in California with Airborne High Resolution Imagery: DAIS
Gong P, Biging G, Yu Q, Clinton N. 2005. Object-Oriented Vegetation Classification at Peninsular National Parks in California with Airborne High Resolution Imagery: DAIS. Center for Assessment and Monitoring of Forest and Environmental Resources, College of Natural Resources, University of California, Berkeley. Berkeley, CA
We conducted vegetation mapping at Point Reyes National Seashore and Golden Gate National Recreation Area in Northern California to create a comprehensive vegetation inventory, covering about 714 km2 and requiring 48 DAIS frames. Information was stored in a vector database comprised of a series of shape file (one per DAIS frame). In this project we evaluate the capability of high spatial resolution airborne DAIS (Digital Airborne Imaging System) imagery for detailed vegetation classification at the alliance level with the aid of topographic data. Image objects as minimum classification units were generated through FNEA (Fractal Net Evolution Approach) segmentation using eCognition software. For each object, 52 features were calculated including spectral features, textures, topographic features and geometric features. After statistically ranking the importance of these features with the CART algorithm (classification and regression tree), the most effective features for classification were used to classify the vegetation. Due to the uneven sample size for each class, we chose a non-parametric (nearest neighbor) classifier. We built a hierarchical classification scheme and selected features for each of the broadest categories to carry out the detailed classification, which significantly improved the accuracy. Pixel-based maximum likelihood classification (MLC) with comparable features was used as a benchmark in evaluating our approach. The object-based classification approach overcame the problem of salt-and-pepper effects found in classification results from traditional pixel-based approaches. The method takes advantage of the rich amount of local spatial information present in the irregularly shaped objects in an image. This classification approach was successfully implemented. Computer-assisted classification of high spatial resolution remotely sensed imagery has good potential to substitute or augment the present ground-based inventory of National Park lands.
- Type
- Unpublished Report
- Authors
- Gong, Peng; Biging, Greg; Yu, Qian; Clinton, Nick
- Date of Issue
- 2005-05-20
- Publisher
- Center for Assessment and Monitoring of Forest and Environmental Resources, College of Natural Resources, University of California, Berkeley
- Units
- GOGA , PORE
- Keywords
- DAIS, Digital Airborne Imaging System, Fractal Net Evolution Approach, vegetation classification, vegetation inventory, vegetation mapping
- Subjects
- Ecological Framework: Biological Integrity | Focal Species or Communities | Vegetation Complex