The climate change leads the disaster occurring in everywhere. Remote Sensing as useful technology can detect the disaster quickly even though in a remote area and bad weather condition. The extreme rainfall occurred on 28-29th August 2008 around Chugoku, Shikoku, Tokai, Kanto in Japan that leading the flood. The most impacted area is Okazaki City and Anjo City. According to Japan Meteorology Agency data, the rainfall was 146.5 mm in one hour. In this research, we utilized pre-disaster and post-disaster images of the ALOS/PALSAR satellite to extract the inundation of flood in Okazaki City and Anjo City. By extracting the backscattering value of each object from a predisaster and post-disaster image, then decided the threshold of the backscattering, we extracted the inundation area. The result was validated by using flood report from Okazaki and Anjo City. This method can be used as a quick response to flood disaster.