Land Cover Classification with Multi-source Data Using Evidential Reasoning Approach

Land cover classification is the core of converting satellite imagery to available geographic data.However,spectral signatures do not always provide enough information in classification decisions.Thus,the application of multi-source data becomes necessary.This paper presents an evidential reasoning (ER) approach to incorporate Landsat TM imagery,altitude and slope data.Results show that multi-source data contribute to the classification accuracy achieved by the ER method,whereas play a negative role to that derived by maximum likelihood classifier (MLC).In comparison to the results derived based on TM imagery alone,the overall accuracy rate of the ER method increases by 7.66% and that of the MLC method decreases by 8.35% when all data sources (TM plus altitude and slope) are accessible.The ER method is regarded as a better approach for multi-source image classification.In addition,the method produces not only an accurate classification result,but also the uncertainty which presents the inherent difficulty in classification decisions.The uncertainty associated to the ER classification image is evaluated and proved to be useful for improved classification accuracy.

Comparison and Effects of Different Climate-Vegetation Models in Areas of Complex Terrain under Climate Change

Identifying the impacts of climate change is important for conservation of ecosystems under climate change, particularly in mountain regions. Holdridge life zone system and Koppen classification provide two effective methods to assess impacts of climate change on ecosystems, as typical climate-vegetation models. Meanwhile, these previous studies are insufficient to assess the complex terrain as well as there are some uncertainties in results while using the given methods. Analysis of the impacts of the prevailing climate conditions in an area on shifts of ecosystems may reduce uncertainties in projecting climate change. In this study, we used different models to depict changes in ecosystems at 1 km × 1 km resolution in Sichuan Province, China during 1961-2010. The results indicate that changes in climate data during the past 50 years were sufficient to cause shifts in the spatial distribution of ecosystems. The trend of shift was from low temperature ecosystems to high temperature ecosystems. Compared with K？ppen classification, the Holdridge system has better adaptation to assess the impacts of climate change on ecosystems in low elevation（0-1000 m）. Moreover, we found that changed areas in ecosystems were easily affected by climate change than unchanged areas by calculating current climate condition.

Description and Evaluation Approach for Uncertainty of RS Images Classi- fication

With the development of researches on the classification quality of remote sensing images, researchers thought that uncertainty is the main factor that influences classification quality. This study puts forward an approach to uncertainty repre- sentation, which is developed from two aspects： formalized description and comprehensive evaluation. First, we complete the classification using fuzzy surveillance approach, taking it as a formalized description of classification uncertainty. Then we in- troduce a hybrid entropy model for classification uncertainty evaluation, which can meet the requirement of comprehensive reflection of several uncertainties, while constructing the evaluation index from pixel scale with the full consideration of the different contribution to the error rate of each pixel. Finally, an application example will be studied to examine the new method. The result shows that the evaluation results fully reflect the classification quality, when compared with the conventional evaluation method which constructs models from unitary uncertainty and category scale.