http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
- 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
- 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
RISS 인기검색어
- 검색키워드
- 저자 : Iisaka, Joji (검색결과 5 건)
- 무료
- 기관 내 무료
- 유료
-
An Automated Method of Man-Made Feature Extraction from LANDSAT TM Data
Iisaka, Joji 대한원격탐사학회 1997 International Symposium on Remote Sensing Vol.13 No.1
This paper describes a method to automatically extract man-made features such as road segments and high-density urban areas from Landsat TM images. The method integrates not only spectral information of manmade objects but also spatial and structural information among spectral objects. It is well known that most manmade objects have simple geometrical dimensions, while other objects in nature show more complex spatial patterns with non-integer dimensions. In this study, fractal information of spatial objects as well as spectral information, is employed to extract road features. Spatial association among spectral objects was applied to delineate a high-level ground cover class for high-density urban areas.
-
A NEW VEGETATION INDEX FOR REMOTE SENSING
IISAKA, JOJI,AMANO, TAKAKO SAKURAI 대한원격탐사학회 1999 International Symposium on Remote Sensing Vol.15 No.1
Global vegetation change is one of major global concerns. Remote sensing images provide an efficient and useful data source to estimate global vegetation covers, and a number of methods have been proposed to estimate them. Among them, the NDVI is one of the mast popular indices, and it is-easy to calculate with simple image computing. However, this index is very much affected by the radiometric environment of sensing such as atmospheric conditions and the sun illumination angle. Therefore, it is not appropriate to apply the NDVI to investigate seasonal changes. This paper discusses these problems and proposes an alternative index, MODVI(Modified Vegetation Index) , that is less affected by radiometric environment changes. An experiment was conducted to compare these two indices using temporal Landsat TM sub-scenes.
-
Automated Landscape Indexing for AVHRR Data
Iisaka, Joji,Amamo, Etsuko,Amamo, Takako Sakurai 대한원격탐사학회 1995 International Symposium on Remote Sensing Vol.11 No.1
Radar backscatter (SAR image) from terrain is influenced by two sets of parameters: 1) physical parameters such as complex dielectric constant of the scatterers and surface topography, and 2) radar parameters such as frequency, incidence angle and polarization. For a set of given radar parameters, the strength of the backscattered field from a soil surface and its statistics are complex functions of the surface irregularity relative to the wavelength and the dielectric constant of soil medium. At first, a method for retrieving appropriate physical parameters from field measurements is studied. Then, the effect of soil moisture and surface roughness on radar backscattering from soil surfaces is presented.
-
A Feature Mixing Model for AVHRR Data
Amano, Takako Sakurai,Iisaka, Joji,Takagi, Mikio 대한원격탐사학회 1996 International Symposium on Remote Sensing Vol.12 No.1
Since the spatial and spectral resolution of AVHRR data is coarse, each resolution cell usually contains many terrain features, not all of which are spectrally distinguishable. Therefore spectral differences between resolution cells often result from the differences in the ratio of the terrain features, rather than differences in the terrain features themselves. The Feature Mixing model assumes that each pixel in a scene consists of a mixture of a very small number of basic spectral features of the landscape, such as vegetation, soil or inorganic materials, and dark objects. The model was tested using AVHRR data to examine the feasibility of using it as a more general measure of monitoring global environmental changes. The results of the preliminary study were positive, although there are still many tasks to be done.
-
Extraction of Linear Features from an AVNIR data of a Suburban City
Amano, Takako Sakurai,Naito, Takahiro,Iisaka, Joji,Takagi, Mikio 대한원격탐사학회 1997 International Symposium on Remote Sensing Vol.13 No.1
Various linear geographic features were extracted from an ADEOS AVNII2 multispectral image of a small suburban city. Spectral data were transformed and normalized into three images representing the proportions of vegetated, inorganic, and dark surfaces, and binarized at multiple threshold levels. Each pixel in each binary image was evaluated for line-likeness at the local level. Values of line-likeness were summed over all threshold levels. The main components of linear features were extracted by applying simple image algebra, morphological operations, and size operations to the resulting images. Additionally, higher spatial operations needed to extract finer details were explored.
연관 검색어 추천
이 검색어로 많이 본 자료
활용도 높은 자료
