A Landsat data classification method to classify land cover using modified unsupervised classification techniques is presented.
It is possible to interpret physical properties of the classification result of this method. This method combines clustering with factor analysis and typical land cover spectral reflectance properties of each band. For the research, a test area including fire damage sites was used. A fire damage condition map of this area, prepared by field surveying and aerial photo interpretation, was already available.
The adopted land cover classification procedure was as follows ; Step 1. (Identification of fire damage representa-tives)
Clustering was carried out on the randomly selected Landsat data for the area of interest. Fac-tor analysis was done using band averages of the clusters which had been gotten by the above mentioned clustering. Factor scatter diagram was plotted with the two factors which were enough to represent the total variance. The factor loading of each band on each factor was examined, and the physical properties of each factor was interpreted. The relative positions of each cluster in the diagram and the spectral curve of each cluster were examined, and the clusters were regrouped into such a smaller number of groups that could be interpreted physically. The clusters belonged to the group which was interpreted as fire damage, were taken as the representatives of fire damaged area.
Step 2. (Estimation of severity of damage) Clustering was carried out on the Landsat data which were only belonged to the selected clusters (fire damaged area) in Step 1. Three severity class groups of fire damaged area was interpreted by the same procedure as in Step 1.
Step 3. (Classification of fire damaged area) The Landsat data of the fire damaged area were classified into three fire damage severity classes by the maximum-likelihood classification method using the three groups which were gotten in Step 2 as the training data. Classification accuracy was about 80% when the classified Landsat image was compared with the existing fire damage condition map.

View full abstract

An algorithm of displaying an image of remotelysensed data with restricted colors has been developed. First, peaks over 256 in three-dimensional histogram space of three channel data are detected, which seem to represent characteristics of the image. Next, the 256 peaks, which correspond to the same numbers of colors in RGB space, are selected among those by the present algorithm. Each pixel in an image is assigned to the nearest neighbor among the 256 colors in Euclid metric. The comparison are shown with the conventional MCA (Median Cut Algorithm) and PA (Popularity Algorithm) methods by using SPOT HRV/XS data. Finally, the recomposed image with 256 colors by the present algorithm maintains the quality of the original data. This is also quantitatively shown in quantum error analysis.

View full abstract

Image registration with registers two or more images in geometric conformance is a fundamental tool of image processing. This is one of important preprocessing of image processing. For example, it is useful in change detection, stereoscopic estimation and mosaicking of several images. This paper describes the registration processing system with user support function to accomplish the high accuracy of geometry and the easy operating of the user interface. Concerning conventional registration processing system, some transformation formulas such as affine or Hermitean transformation are installed, but they are not adaptive transformations for distortion characteristics of image, So it needs the trial and error to define a transfomation formula corresponding to the distortion characteristics of the image. Then we have to select GCPs (Ground Control Point) and to input them to image processing system. In general, it is difficult to select many GCPs from satellite image, so hardly can we select sufficient number of GCPs for the registration and its estimation, and the most time-consuming part of the registration processing is the computation of the interpolation processing. Therefore we develop
image registration processing system from following two points for user interface to process registration efficiently under the interactive computer environment and the high accuracy correcting process as follows Simplification of GCP selection ; Juxtaposition of two images, input of GCP coordinates by indication devices, e.g., track ball and automatic detection of the candidate GCP. (2) Automatic estimation of distortion model ; Modeling of distortion by polynomial and defining an optimum polynomial order automatically by PSS (Prediction Sum of Square). As a result of estimation of the performance of developed system, the registration accuracy of about less than 1 pixel are obtained and it becomes about one-forth comparing as conventional system to select and input GCPs. Therefore we can develop the user supporting algorithms of less operation load for a high accuracy image registration processing.

View full abstract

A new classification method has been produced to decrease the computing time of multispectral satellite data classification analysis, maintaining the accuracy as precise as the Maximum Likelihood Method (MLM). In this method, the dimensions of the original data are arbitrarily changeable, ie., M * N original data reduce to M'* N' cells which consist of L * L pixels (L=M/M'=N/N'). The dimensions of training areas are usually chosen to be the same as that of a cell.
The similarity between a cell and a training area is defined as the degree of the overlap of both histograms. This idea is analogous to the Fuzzy theory in a sense. The effect of the cell size and histogram smoothing on classification was carefully discussed in connection with the MLM. It has been found that this method (called Histogram Overlay Method, HOM in abbreviation)results in fair agreement with the MLM.
HOM was applied to the investigation of the turbidity on the Kanmon area between Shimonoseki and Kitakyushu, and reasonable results were obtained. In general, the variation of CCT counts in the ocean is small in comparison with those on land, so that "cell by cell" classification would be sufficient. In addition to this, HOM can be used on personal computers since it takes much less time in classification compared to MLM.

View full abstract

It is very important to obtain the information on soil physical properties and the growth condition of trees, when we make runoff analysis in forest land. But actual vegetation map is not including the growth condition of trees, and it is difficult to obtain the information on the soil. Therefore, it is thought that the usefulness of satellite data have to be verified in this aspect.
This paper aims at estimating the ratio of coarse pore, which relates to hydrologic cycle in forest using satellite data and applying it to flood-runoff analysis. In mountainous terrain, it is necessary to eliminate the terrain effect to radiometrical distortion for satellite data. Therefore, the radiometrical distortions of used LANDSAT/TM data were eliminated using digital terrain model (DTM). Many measurement points were selected in the area for survey of various physical soil properties and vegetation conditions of the forest.
In order to estimate the ratio of coarse pore from satellite data, we made linear multiple regression equation through multiple regression analyses on the relationship between the ratio of coarse pore and corrected LANDSAT/TM data to each of categorized three forest types. As the result, it was found that the ratio of coarse pore related to the band ratio of near-infrared/red. Further, we tried to apply the estimated ratio of coarse pore to runoff simulation model. It was found that this method was useful to estimate the ratio of coarse pore, because of the simulated hydrograph well coincide with the observed one.

View full abstract

In the fields of cryospheric sciences, there is a wide variety of research topics related to satellite remote sensing of ice. Major requirements of these researches are to derive changes in the surface features, extent and thickness of various ice, such as continental ice sheets, glaciers and sea ice, from satellite informations. The present status and future problems of the remote sensing of ice are reviewed as an introduction of this special issue.

View full abstract

The measurements of dielectric constants of ice were made at 10 GHz with a standing wave method using Antarctic core ice, Mendenhall Glacier ice and artificial ice. The obtained results are as follows. The relation between permittivity, ε', and ice density, p, is given by ε'=2.8p+0.5. The loss factor, ε", is not influenced by grain-boundaries and/or dislocations but by impurity concentration. The temperature, T, dependence of conductivity, σ, is given byσ=σ₀ exp(-E/RT). In the case of artificial ice, E=31 kcal/mole at temperatures higher than -8^°C and E=6.2 kcal/mole at temperatures lower than -8^°C.
Based on these results, the importance of the basic study in the problems on remote sensing in cryosphere is discussed.

View full abstract

Remote sensing of ice by radio waves, especially by spaceborne microwave remote sensor is reviewed. Usefulness of microwave radiometers and synthetic aperture radars for the remote sensing of sea ice is shown. Also, the remote sensing of continental ice by the microwave altimeter and synthetic aperture radar is introduced. Finally some subjects
for the future remote sensing of ice from space in Japan are discussed.

View full abstract

Glaciological investigations by radio echo sounding and satellite remote sensing technique have been carried out in the area of glaciers, Greenland and Antarctic ice sheets.
Airborne radio echo sounder is a sophisticated instrument to measure ice thickness of glacier and ice sheet in the polar region. The radio echo sounder newly designed at 179 MHz indicated that the maximum ice thickness observed was about 2, 800 m in depth and the Z-scope data showed the fine multiple internal layers in the ice sheet. In addition, an estimation methods were proposed for radio scattering characteristics of the bottom of ice sheet along Shirase glacier with the subglacial water and also proposed for the accurate location of the grounding line in Roi Baudouin Ice Shelf.
To gain the perspective that includes the dynamic features of the ice sheet, it is neccessary to view the ice sheet from rpace with enhanced digtal imagery. Many of the most significant dynamic features of ice sheets can be identified by a careful examination of AVHRR imagery. The relatively low resolution of this instrument makes it ideal for obtaining a broad view of the ice sheets, while its wide swath allows coverage of areas beyond the reach of highresolution imagers either currently in planned orbit.

View full abstract

MOS-1 has three different sensors on-board, which are MESSR, VTIR and MSR. This paper discusses about the utility of MOS-1 sensors for monitoring of ice sheet and sea ice.
Firstly, the movements of Shirase glacier and the calving of ice sheet in Antarctica was analyzed using MOS-1 MESSR data (IFOV : 50 m) of 1989 and Landsat TM data (IFOV : 30 m) of 1988. The result showed that the tongue of Shirase glacier moved approximately 2.7 km and the edge of ice sheet in this area moved about 500 m in 1988-1989.
Secondarily, VTIR images (IFOV : 0.9 km for visible band, 2.7 km for thermal bands) of the Sea of Okhotsk were analyzed, which described the detailed pattern of sea ice distribution in this region. However the cloudy weather of this region reduces the usefulness of VTIR data for ice monitoring. On the other hand, low resolution (32 km for 23 GHz, 23 km for 31 GHz) MSR data has good atmo-spheric penetration in cloudy weather. The coastline of the Main Island of Japan was clearly captured through heavy clouds in the MSR image of Feb. 9, 1988. Finally, simultaneously collected VTIR and MSR data were merged and the multi sensor data set was composed. MSR data analyses of estimating ice concentration, water vapor amount or liquid water contents can be reinforced or verified by using this data set. The false color composite images of visible band (VTIR ch. 1), thermal band (VTIR ch.3) and micro-wave band (MSR ch.1 or ch.3) suggest the utility of VTIR & MSR multi-sensor data set for ice monitoring.

View full abstract

Radio echo sounding is most useful to understand Antarctic ice sheet and ice shelf. We had previously reported the method to infer roughness of ice bottom from A-scope shapes. The back-scattering coefficients of ice sheet surface and bottom were calculated using developed radar equations. Roughness was inferred from the relationship between scattering coefficient and incidence angle.
Smoothness of the ice shelf bottom was inferred from the coherent component of scattering characteristics. Coastal regions have more smooth bottom features than near continental areas. This result suggests that the interaction between sea water and ice shelf bottom associated with the distance from the grounding line. Furthermore, at the coastal edge of the shelf where echoes from the bottom show very specular scattering characteristics, double trip scattering echoes were observed. This suggests that power losses due to specular scattering are few.
In addition, the bottom crevasses were observed at the limited region.

View full abstract

The Communications Research Laboratory (CRL) has been developing new radar systems in cooperation with the National Institute of Polar Research (NIPR) to support research activities and transportation in Antarctica area. A UHF step frequency radar system to measure sea ice thickness and a short pulse C-band radar to detect hidden crevasse were developed. These radio wave systems must have high resolution at close range. They not only have to penetrate into snow and ice, but also have to receive echoes which indicate where exist
boundary layers between ice and water for a sea ice thickness measurement or between ice and air for a crevasse detection. The characteristics of the instru-ments and results of preliminary experiments to confirm the characteristics of the radars are shown.

View full abstract

Ice monitoring is one of the most important monitoring items of earth environment problems because ice conditions closely relate to earth warming. Drift ice in Sea of Okhotsk is very important for Japan and it is necessary to study these ice conditions. NASDA has conducted MOS-1 airborne experiment during 1984 FY and 1985 FY, and MOS-1 Verification Program (MVP) experiment during 1987 FY and 1988 FY. As a part of MOS-1 airborne experiment and MVP, NASDA conducted airborne experiment in Hokkaido during Jan. 25, 1985 and
February 9, 1988 in order to evaluate effectiveness of MOS-1 data in the field of ice monitoring. In case of MVP, experiment was conducted on the same date as MOS-1 passing over test site.
In this paper, outline of these experiments and their results are presented.

View full abstract

For observing in detail the distribution of drift ice off the Okhotsk Sea coast of Hokkaido, a radar network was constructed in 1966-1968 by Sea Ice Research Laboratory, Hokkaido University.
This radar network allows a continuous monitor-ing of real-time ice field scenery along a 250 km coastline as far as about 60 km into the Okhotsk Sea. A series of 20-years long ice distribution records from 1969 to 1989 was obtained by this radar newtwork.
Based on these data,
1. The dates of appearance and disappearance of sea ice in the radar images, and the number of days of sea ice existing in the radar images were discussed statistically.
2. The relations between the ice amount off the Okhotsk Sea coast of Hokkaido and the air temper-ature along the shore were discussed.
3. Variations in ice amount off the Okhotsk Sea coast of Hokkaido were compared with those in the Sea of Okhotsk.
4. The relations between the variations in ice amount off the Okhotsk Sea coast of Hokkaido and the variations in the atmospheric pressure gradient were discussed.

View full abstract

Sea ice monitoring from space is overviewed. The relationships among the albedo of the earth, hydrology cycle, ocean dynamics, atmospheric physics/ dynamics, sea ice, in-land ice, open water and polynya are introduced together with the brief, descriptions of the basic principles for microwave radiometer, scatterometer, synthetic aperture radar, laser atmospheric sounder/altimeter, visible and near infrared radiometer. Further sea ice monitoring in the Earth Observing System (EOS) Program including the background and the plan for experiments is also introduced.

View full abstract

The purpose of this report is to investigate character of MOS-1 MESSR data by visual interpretation of photographic images made with imaging method suitable for visual interpretation of TM and MSS images. Sea and land area were examined separately because of difference of data character. On sea area data character was not complicated and then MESSR and MSS images were simply compared. On land area detailed examination was made on two scene data of MESSR because there were various kinds of objects and data were very complicated. On scene 1, all possible color composites were prepared from MESSR data and influence of band combination and color sequence on visual interpretation was examined. On scene 2, three kinds of B/W images were made from MESSR data, and also TM and MSS images of the same area to MESSR were prepared to compare mutually by visual interpretation on various kinds of objects. It is found that these MESSR images are not so suitable to visual interpretation on land area because these data ranges are very narrow and data levels
are located low. However it is confirmed that these images are useful to sea area.

View full abstract

The changes in land coverage on the Suo district, Yamaguchi Prefecture, were observed by MOS-1/ MESSR to trace back the process of its industrial development and predict analogous cases in Northen-East Asia.
The salt fields in the Seto Inland Sea were almost all replaced with industrial sites in the course of the intensive growth age during 1960s and 70s. The satellite observations have revealed that the salt fields in the Tokuyama 'and Kudamatsu areas, southern half of Suo, are exclusively used for indus-try. It is also found that some of the salt fields in the Mitajiri and Yamaguchi areas, the western half of Suo, are still left to be industialized. The Kunsan area of South Korea was chosen as a representative case. By overlaying an old map onto the recent LANDSAT TM image, the changes in the past half century were examined. The successive changes in the geographical features were also
investigated using the time series LANDSAT data from 1979 to 1987. It revealed that there are significant changes in land aspects as well as coastal lines due to the consecutive reclamation work.
The capability of the Kunsan area as an industrialized site was discussed from various angles in connection with the Japan's experience regarding the industrial development in the Seto Inland district.

View full abstract