Publications Search

We use crowdsourced GIS data from OpenStreetMap to help automate the mapping of land cover in Landsat satellite imagery. The OpenStreetMap data had limited area coverage, but was useful for building a model which could classify land cover in the other areas.

We developed a method for mapping residential areas using freely available satellite imagery. The resultant residential area map can be used for risk assessment (e.g. to estimate the population living in a flood-prone area). Our method can be applied for flood risk assessment in other watersheds in the future.

We analyzed the change in vegetated and urban (impervious) area in the Silang-Santa Rosa sub-watershed of the Philippines to assess the implications related to flood risk. Our main findings were: -The impervious area of sub-watershed increased by 54% (from 3,239 ha. to 4,988 ha.). -The vegetated area decreased by 21% (from 8,509 ha. to 6,760 ha)....

This dataset can be used by researchers to test different land use/land cover mapping methods using crowdsourced geographic data and Landsat satellite imagery. The study area is the Laguna de Bay area of the Philippines, and the satellite data is from 2014-2015. This dataset will also be available on the "UCI Machine Learning Repository" (https:/...

We developed a fast method for land cover mapping using satellite imagery and crowdsourced geographic data (land use polygon data). The method is less time- and labor-intensive than traditional land cover mapping methods because the crowdsourced data substitutes for training data collected the traditional way (through field survey or visual image...

CO2 emissions from expected land use change were calculated using GIS techniques. ~500,000 tons CO2 emissions expected by 2025 under the BAU land use plan.