Establishment of the watershed image classified rule-set and feasibility assessment of its application

Extreme weather events result in catastrophic disasters around the world recently. More commonly, these disasters occurred due to multiple reasons that coupled together. To develop disaster prevention strategies, application of satellite images can be effective because of its promptness and vast coverage area. This study established an image classified rule-set which used the object-based image analysis methodology. There are several surface features in the watershed image that were classified based on the proposed rule set, including main channels, secondary channels, sandbars, alluvial fans, landslides and place of the geotechnical damage. The study applied this rule-set in different watersheds and different-resolution satellite imagery, and assessed the feasibility by comparing in-situ data and calculating the error matrix. The results showed that the rule-set logic can be flexible in different watersheds and different images. The classification of the rule-set is reproducible and accurate. As this result we can apply the rule-set to disaster management and land use planning in the future work.