All issues
Volume 14 (1994)
- Issue 4 Pages 285-
- Issue 3 Pages 203-
- Issue 2 Pages 111-
- Issue 1 Pages 3-
Volume 14, Issue 2
Displaying 1-9 of 9 articles from this issue
- |<
- <
- 1
- >
- >|
-
Hirosi MURAI, Masahiro OKAMURA, Sigeru OMATU1994 Volume 14 Issue 2 Pages 111-121
Published: June 30, 1994
Released on J-STAGE: May 22, 2009
JOURNAL FREE ACCESSNeural network architecture enables us to attain flexibility for pattern recognition in remote sensing. In this paper, we propose a pattern recognition system of remote sensing data using two kinds of neural networks. It uses both self-organizing and back-propagation (BP) methods. Although the BP method can easily train the network, it is difficult to find exact training data. The self-organization method can calculate connection weights autonomously and select the training data easily. But we must prepare a large size network for input data.
In order to improve ability of pattern recognition, we divide the pattern recognition system into three subsystems. Each of them has been processed in a sequential way. The first one clusters the remote sensing data into some regions based on Kohonen's feature map. At the second step using the clustered images and geographic knowledge, training data sets are classified into many categories based on the BP method. The third step corrects miss-classified pixels by using IF THEN rules. The rules are described on the basis of human criterion for pattern recognition. Effectiveness of the proposed method is illustrated by simulation results of remote sensing TM data for checking data sets observed around Tokushima City in Japan.View full abstractDownload PDF (3591K) -
Ikuo HORIGUCHI, Dali LUI, Hiroshi TANI, Takashi MACHIMURA1994 Volume 14 Issue 2 Pages 122-131
Published: June 30, 1994
Released on J-STAGE: May 22, 2009
JOURNAL FREE ACCESSA forest fire occured on the border between China and Russia in 1987. The burnt area in China was calculated to be 11, 744 km2 and the total burnt area (in China and Russia) 69, 772 km2 using NOAA/AVHRR data.
Regeneration of vegetation in burnt areas was analyzed using NOAA/AVHRR data for 1986-1991. The burnt areas were classified into three areas: severely burnt areas, moderatly burnt areas and lightly burnt areas. The Vegetation Index, surface temperature and regression coefficient between Vegetation Index and surface temperature were calculated for those three burnt areas and non burnt areas (forest areas). Seasonal and yearly changes of the Vegetation Index, surface temperature and regression coefficient for 1986-1991 were analyzed.
It was confirmed by yearly changes that the Vegetation Index increased, the difference in temperature between burnt area and forest area decreased and the absolute value of regression coefficient increased. It was possible, therefore, to estimate the regeneration period for burnt areas using yearly change of the Vegetation Index, surface temperature and regression coefficient. The regeneration period for the severely burnt area were estimated to be ten years by the Vegetation Index, thirteen years by surface temperature and sixteen years by regression coefficient.View full abstractDownload PDF (728K) -
Kazuhiko MASUDA, Tsutomu TAKASHIMA1994 Volume 14 Issue 2 Pages 132-142
Published: June 30, 1994
Released on J-STAGE: May 22, 2009
JOURNAL FREE ACCESSThe derivation of aerosol parameters using polarimetric measurements by the airborne version of POLDER (POLarization and Directionality of the Earth’s Reflectance) over the Mediterranean sea is described. The degree of polarization of the upwelling radiation at the altitude of the airplane is computed by the doubling-adding method using a realistic atmosphere-ocean model. By comparison with measurements, characteristics of the aerosols over the ocean are inferred. The degree of polarization for the lunge size distribution with v=3.5 for small optical thickness, viz., τaer≤0.1, shows features most similar to the measurements. The simulation indicates that the wind direction has little effect on the degree of polarization of the upwelling radiation, whereas whitecaps greatly reduce the degree of polarization. In the solar plane, the computed degree of polarization agrees well with the measurements for τaer≤0.1 if the effect of whitecaps is excluded, whereas it is several percent smaller than the measured values near the backscattering direction. The computed minima of the degree of polarization are located at θ=38°, closer to the nadir than in the measurements.View full abstractDownload PDF (2465K) -
Masanobu SHIMADA1994 Volume 14 Issue 2 Pages 143-154
Published: June 30, 1994
Released on J-STAGE: May 22, 2009
JOURNAL FREE ACCESSSince the first activation of JERS-1's Synthetic Aperture Radar (SAR) on beginning of April 1992, calibration efforts for NASDA's SAR products are being conducted by using calibration instruments with known signal intensities and locations, such as 2.4 meter trihedral corner reflectors and active radar calibrators. Data analysis result shows that the conversion factor from Digital Number (DN) of SAR product to Normalized Radar Cross Section (NRCS) of the target is constant over full image swath and is determined as -68.51 dB with the accuracy of ±1.86 dB (1 sigma) and its long term stablity is within ±1.8 dB. The image quality evaluation conducted in parallel also shows that the image quality is as good as designed.View full abstractDownload PDF (4419K) -
Hong-Min SON, Ikuo ARAI, Tsutomu SUZUKI1994 Volume 14 Issue 2 Pages 155-163
Published: June 30, 1994
Released on J-STAGE: May 22, 2009
JOURNAL FREE ACCESSIn case of real aperture microwave radiometers, it is required a large antenna to realize a high spatial resolution. But it is impractical to load a satellite with large antenna.
It is well known that the beam compression by correlation arrays for radars can reduce the beam width equivalently without increasing antenna aperture size. The narrow beam obtained by the beam compression techniques is attractive but the first side-lobe level is very high.
It is also required for microwave radiometers that the directional characteristics has low side-lobe as well as narrow beam width, because a high side-lobe level causes the errors in measurements of antenna temperature.
In this paper, it is proposed a new signal processing method for microwave radiometers to reduce the first side-lobe level as well as beam width. Narrow beam width with low side-lobe is realized by combining the beam compression techniques for radars and aperture synthesis.
A L band prototype radiometer was constructed and its responses to both one and two noise sources were measured.The results showed that 1/2.8 beam compression ratio and low first side-lobe level of -25 dB had been achieved sucessfully.View full abstractDownload PDF (493K) -
[in Japanese]1994 Volume 14 Issue 2 Pages 164-169
Published: June 30, 1994
Released on J-STAGE: May 22, 2009
JOURNAL FREE ACCESSDownload PDF (2859K) -
1994 Volume 14 Issue 2 Pages 173-175
Published: June 30, 1994
Released on J-STAGE: May 22, 2009
JOURNAL FREE ACCESSDownload PDF (177K) -
1994 Volume 14 Issue 2 Pages 176-180
Published: June 30, 1994
Released on J-STAGE: May 22, 2009
JOURNAL FREE ACCESSDownload PDF (1381K) -
1994 Volume 14 Issue 2 Pages 195
Published: June 30, 1994
Released on J-STAGE: May 22, 2009
JOURNAL FREE ACCESSDownload PDF (77K)
- |<
- <
- 1
- >
- >|