ROBUST BUILDING FOOTPRINT EXTRACTION FROM BIG MULTI-SENSOR DATA USING DEEP COMPETITION NETWORK

This paper represents a proper input by combining temperature, fault and time for different thermal anomaly detection methods and then using them for estimating earthquake intensity. Results show that the proposed input data has better outcome in each of the thermal anomaly detection methods compared to the ordinary input data and the estimated intensities of most earthquakes are very close to the actual intensities. This approach can be used as a tool for future unknown earthquakes prediction.