When a regional information database is configured, managed and used, it is one of the important maintenance operations to update the attribute data. Main attribute data of the regional information database is classified into settled, semi-settled, and unsettled types as viewed from the updating frequency. This classification makes itpossible to reduce update work by handling the data in time-series; however, this causes data redundancy. This problem can be solved by normalizing data tables and utilizing data relationship. This study attempted the configuration of the regional information database using a relational database manegement system based on the classification to improve the database management that requires a long time and much labor. The study was proposed as one of the techniques that manage the attribute data efficiently.
Accurate geometric correction including ortho-rectification is considered for Landsat TM imagery of large rugged terrain. Ortho-rectification requires off-nadir distance of satellite image and accurate digital elevation model. The digital elevation model published by the National Geographycal Survey are reprojected on the UTM coordinate system of the concerned region (Iwate Prefecture) with 30m spatial resolution. A simple method for estimating the offnadir distance are proposed based on the observation of the TM image which is normally processed by NASDA. For precise geometric transformation, 120 ground control points are systematically selected, and their image coordinates and map coordinates are carefully measured. Statistical analysis of the obtained points reveals that the errors both in pixel and line positions are less than 0.7 pixel if the relief displacement is removed, otherwise the error in pixel position increases by 0.3 pixel. These errors can be statistically explained by the pixel identification error (0.5 pixel), the digitizing error (0.3 pixel), and the relief displacement (0.6 pixel) . The ortho-rectified image is consistently overlaid on the map image.
The OLS (Operational Linescan System) sensor on board the DMSP (Defense Meteorological Satellite Program) can observe the lights in the night. A nighttime OLS visible-near infrared (VNIR) channel image was overlaid on the simultaneously corrected OLS thermal infrared (TIR) channel image for the area around Japan. The OLS composite image showed a clear relationship between the location of fishing fleet lights detected by the VNIR channel and the sea surface temperature (SST) distribution observed by the TIR channel. Many fishing fleets were located at the cold side of boundary area of warm current and cold current. Since some types of fishes are likely to gather in certain sea temperature zones, the OLS composite image may provide useful information for the fishing fleets monitoring as well as for the marine resources management.
Unusual weather during the summers of 1990's was detrimental to forests and crops. The scientists responsible for the observed forest decline have not been identified, and still trying to find the causes. In this study, measurements of leaf spectral reflectance, photosynthesis, and stomatal conductance of Stewartia pseudocamellia Maxim. (Natsutsubaki) seedlings were carried to detect an indicator of plant water status. Reflectance on infrared wavelength increased with decreasing leaf water content. The slope and inflection point position of the red edge reflectance feature was determined by calculating the first derivative of the spectral curve. A distinctive blue shift, defined here as a shift of the red edge inflection point to shorter wavelengths, is seen in reflectance data from samples collected of the water deficient seedlings. Stress index (SI) was also derived from leaf spectral reflectance data. The photosynthesis and stomatal conductance were affected by low water potential, and these parameters decreased with increasing SI value. The relationships between the physiological parameters of tree seedlings and SI derived from remote sensing data suggest that SI value is useful for detecting and monitoring stress status of forests.
The Airborne Multi-Spectral Scanner (AMSS) is an airborne scanning spectrometer which provides spectral information similar to that which will be provided by the Global Imager (GLI) scheduled to be launched on the ADEOS-II platform in 2000. The AMSS acquires high spatial resolution imagery in the range of 0.40 to 12.45 microns, and a cross track scan width of +/- 35 degrees. This paper introduce the outline of AMSS and the result of image quality check using Integrating Sphere and Black body. As a result, it was confirmed that the data of visible and near-infrared region has yet meet to the specifications and insufficient for scientific use.
Large part of urban area of Kobe, Ashiya and Nishinomiya city were damaged by the 1995 Hyogoken-nanbu earthquake. The authors attempted to detect their damages by ERS-1/SAR intensity images. The intensity images obtained before the earthquake indicated urban structure clearly, however, the structural pattern was unclear in the images after the earthquake. Norm type correlation coefficients of data pair between before and after the earthquake indicated decorrelation at damaged urban area. The decorrelation by the earthquake was detected obviously in comparison with the correlation of the data pairs which interval did not include the earthquake occurrence. The areas where the backscatter correlation changed significantly before and after the earthquake correspond to damaged built-up areas observed by aerial photographs. The ground surface change by the earthquake, such as collapse or burnt down of buildings, should cause the change of backscatter characteristics and further study is necessary to investigate its mechanism. Collapsed built-up area has not been recognized by optical sensor images like SPOT/HRV. The result of this study suggests the usefulness of SAR intensity images for urban disaster observation.
An image data compression method based on the compression (DCT: Discrete Cosine Transformation) featuring an image restoration method is proposed. DCT image compression is widely used and has four major image defects. In order to reduce the noise and distiortions, the proposed method expresses a set of parameters for the assumed distortion model based on an image restoration method. The results from the experiment with Landsat TM (Thematic Mapper) data of Saga show a good image compression performance of compression factor and image quality. Namely, the proposed method achieved 25% of improvement of the compression factor compared to the existing method of Discrete Cosine Transformation with almost comparable image quality between both methods.