The early detection of land cover change can play a significant role in early warning and solution of various environmental problems. Frequent monitoring is required to realize the providing of stable warning information. Advantage of the SAR image (i.e. weather free observation) is utilized by accumulating these stably observed time-series images in the developing method. The method consists of the following functions, (1) extraction of the temporal-profile of every pixel, (2) extraction of the time-series change pattern using time-series analysis, (3) prediction of the future DN value from systematic part, (4) change determination by comparison between the predicted value and the observed value, and (5) extraction of the spatial extent of the land cover changes. The method was examined on the JERS-1/SAR images (1992.9-1998.9) in the test area. As a result, the method succeeded in providing clear image to support the interpretation of land cover, and in achieving the early detection of the land cover changes.
In this research, we propose an automatic 3D mapping system based on multi-sensor integration that is a combination of IMU (FOG), GPS, laser scanner and digital camera. An automatic mapping means that direct geo-referencing is achieved automatically without any ground control points, which accurately measures ground coordinate values. 3D model is reconstructed by laser scanner as 3D point cloud data, while texture is acquired by digital camera from the same platform simultaneously. The accurate trajectory of the platform with attitude changes is determined through the integration of GPS/IMU and digital camera. All the measurement tools are loaded on the unmanned helicopter, RPH2, which is made by Fuji Heavy Industries Ltd. A method of 3D mapping by integrating all the sensors from an unmanned helicopter is focused in this paper.
The purpose of this study is a monitor of Senjyogahara wetland by multitemporal LANDSAT TM/ ETM+ data. Four indices of NDVI, MWCI, Cv, and Cw were used for the analysis. Cv, and Cw are indices calculated by the pattern decomposition method. The principal component analysis (PCA) and the change vector analysis (CVA) were used for the time series analysis. As a result of the PCA, the principal ingredient in which the change in the period was shown about four indices was able to be obtained. The change of 15 years in the marshland was clarified by the result of those principal ingredient and CVA.1) Senjyogahara region is made dry.2) Odashirogahara region is made moist.3) Sakasagawa-river inflow part is made dry.4) The water current dividing into parts with the national road has gone. The above-mentioned tendency was able to be confirmed.