This paper deals with the way of adjustment radiation quantity to develop a unified evaluation method of multi-temporal data as part of an environmental evaluation method using the satellite image. The proposed method is intended to automate reading of training data used for radiation quantity adjustment by extracting common class area during image, following the principle of post-classification comparison, based on land cover classification map produced under the supervised maximum likelihood classification method. Also, it adopted the way of adjusting the frequency distribution of the brightness information and that adjustment is made with threshold value. The effectiveness was shown as follows; it used LANDSAT/TM image. The difference occurring among multi-temporal data decreased, and it proved that the reference image and the comparison became possible relatively.
The acreage estimation of the riceplanted fields is an important subject for agricultural administration, and it requires the high precise measurements in actual work. In order toimprove the accuracy of estimation, high ground resolution satellite data are required in conventional methods. However, there are problems in frequent use of high ground resolution satellite data due to its narrow observation area, requiring a fair weather at time of acquisition and high cost; there are some limitations to use them operationally. A method of referring agricultural plot outline data was developed in this study, which improves the estimation accuracy of the rice planted fields using medium ground resolution satellite data such as LANDSAT-TM, SPOT-HRV or ASTER-VNIR. In this method, the satellite raster image data are not used for totalization, although they are used to discriminate whether the fields are rice planted or not. The acreage of the rice planted fields discriminated as rice planted fields are totalized referring the existing or preliminary measurement data of agricultural plot outline data. By employing the developed method, the acreage estimation of the rice planted fields using remote sensing techniques is remarkably progressed in practical use. This paper described the result which the method was applied to ASTER-VNIR image and the estimation data was compared with the surveyed data. The estimation error ratio was -5.5%.
Forests play an important role in maintaining environmental conditions suitable for life on the earth. Forests have seasonal cycles and change from year to year. External factors such as harmful insects may cause damage, and the surrounding conditions are not constant. All these factors have an effect on the tree crown, therefore crown size measurement is an important task in forest management. The goal of this study was to measure the distribution of the tree crown size in the forest including closed canopies by image processing methods. We assumed that one crown consists of leaves and branches of similar color and that there are shadow areas around the crowns. Additionally each crown grows differently and has different structure. These features are used to identify a circular crown area. The size of a circle representing the crown is computed using the brightness distribution and local fractal dimension within the circular area. To measure the crown size distribution, forest was photographed from a helicopter, and the images were analyzed using the suggested technique. Comparing the analysis results and visual inspection of the same area, we concluded that the proposed technique was effective for measuring the crown size distribution.
ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer) is an imaging instrument that is flying on Terra, a satellite launched in December 1999 as part of NASA's Earth Observing System (EOS) . ASTER is a cooperative effort between NASA and Japan's Ministry of Economy, Trade and Industry (METI) and the Earth Remote Sensing Data Analysis Center (ERSDAC) . ASTER data is distributed by HDF-EOS format, which has a very complicated data structure. It result in the difficulties of handling the data. However, that includes latitude and longitude values of footprints every 6km owing to the precise orbit tracing system with the star tracker. Precise geometric correction is one of the most difficult and indispensable process in remote sensing data. In this study, firstly, a free software is developed to achieve map projection conversion from UTM to plate carriee coordinate with ASTER data in HDF-EOS format. Secondly, map projection conversion is applied to sixteen scenes of ASTER data over Japan Island. Thirdly, accuracy of system geometric correction is evaluated with the castal lines and administrative boundary lines in vector format by visual interpretation. As a result, the root mean square geometric error is 79.2m in North West direction.