14:30 〜 14:45
[HTT14-04] UAS Surveys for Identifying Vacant Houses using Night Light and Thermal Information
★Invited Papers
キーワード：ドローン、熱赤外、夜間光、リモートセンシング、空き家
Changes of the social environment such as the population decline has led to the increasing number of vacant houses in Japan. According to the Housing and Land Survey conducted by the Ministry of Internal Affairs and Communications, approximately 13.6 % of total housing units are vacant across Japan. Japanese government has enforced a new law, regarding the special law about the promotion of vacant houses. The law encourages the local governments to identify and record the situation of the vacant houses. However, the current situation of the identification method relies only on direct observation of the houses, estimating if it is vacant or not. The approach takes massive time and labor costs, which is inefficient to conduct such survey for the whole city and frequently. Situation always changes for house rents, people’s activity, movements, etc., and difficulty rises to grasp the condition up to date.
This work considers the utilization of the Unmanned Aerial Systems (UAS) or hereafter drones, to spatially survey the necessary sites and identify the houses if they are vacant or not. Multiple aerial imageries were collected at different time of the day with a RGB camera. The Structure from Motion technique develops the 3D model of the residential areas which can be used to check through the 3D space. Nighttime flight was also conducted to collect the aerial imageries at nights. The data is processed to spatially identify the night lights that is observed from each home, which can be a direct measurement to identify the human activity. Finally, the thermal information simultaneously collected through the aerial survey was also image processed to observe the heat emission of each home, which also can be an indicator for human activities, hence, it would imply that vacant houses should have less heat transfer. We would like to present and discuss our proposed method for collecting the data and interpreting the developed data for understanding the situation.
This work considers the utilization of the Unmanned Aerial Systems (UAS) or hereafter drones, to spatially survey the necessary sites and identify the houses if they are vacant or not. Multiple aerial imageries were collected at different time of the day with a RGB camera. The Structure from Motion technique develops the 3D model of the residential areas which can be used to check through the 3D space. Nighttime flight was also conducted to collect the aerial imageries at nights. The data is processed to spatially identify the night lights that is observed from each home, which can be a direct measurement to identify the human activity. Finally, the thermal information simultaneously collected through the aerial survey was also image processed to observe the heat emission of each home, which also can be an indicator for human activities, hence, it would imply that vacant houses should have less heat transfer. We would like to present and discuss our proposed method for collecting the data and interpreting the developed data for understanding the situation.