Development of Forest Information Measurement System by Multi-Wavelength and Multi-Polarization High-Resolution Synthetic Aperture Radar

多波长多偏振高分辨率合成孔径雷达森林信息测量系统研制

基本信息

批准号：
17360194
负责人：
TAKAGI Masataka
金额：
$ 6.46万
依托单位：
Kochi University of Technology
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (B)
财政年份：
2005
资助国家：
日本
起止时间：
2005 至 2006
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-17360194/
关键词：
measurement engineering remote sensing forestry imaging radar algorithm

项目摘要

The purpose of this research is to develop techniques to extract forest information from the high-resolution multi-wavelength and multi-polarization synthetic aperture radar, Pi-SAR, developed jointly by the Japan Aerospace Exploration Agency (JAXA) and the National Institute of Information and Communications Technology (NICT). The test site is the coniferous forests in Tomakomai, Hokkaido. First, regression analyses are carried out between the Pi-SAR images acquired in November 2002, and the forest data measured simultaneously on the ground and the forest data measured in August 2003. Using the conventional technique that utilizes the radar cross section (RCS), it is found that there is no significant correlation between the X-band RCS and forest parameters at all polarizations. The L-band RCS, however, is found to increase with increasing forest biomass up to approximately 40 tons/ha. These trends are similar to those already reported by several researchers. Next, because the high-re … More solution SAR images appear to show the structures of the forests, the relation between the image texture and forest information is sought. As a result, the image amplitudes are found to obey the K-distributed probability density function; and that strong correlation exists between the order parameter of the K-distribution in the cross-polarized images and the forest biomass. Further, it is found that the order parameter increases with increasing biomass up to around 100 tons/ha which is well beyond the saturation limit of the conventional RCS method. From the regression curve, the biomass values of unknown forests is estimated and compared with those measured on the ground in 2005-2006. The comparison yields the model accuracy of 86%. Finally, the regression model is updated using all biomass data measured on the ground. This model is considered to be effective for estimating the biomass of coniferous forests on flat ground in the entire areas of Hokkaido; and the accuracy of estimating the forest biomass can be improved to much higher levels by combining the conventional RCS technique and the texture analysis developed in this study. Less

这项研究的目的是开发从高分辨率多波长和多极化合成孔径雷达 Pi-SAR 中提取森林信息的技术，该雷达由日本宇宙航空研究开发机构 (JAXA) 和国家信息研究所联合开发。试验地点为北海道苫小牧市的针叶林，首先对2002年11月获取的Pi-SAR图像与森林数据进行回归分析。地面测量数据和2003年8月测量的森林数据同时进行。利用传统的雷达散射截面（RCS）技术，发现X波段RCS和森林参数在所有偏振态下均不存在显着相关性。然而，随着森林生物量增加到大约 40 吨/公顷，L 波段 RCS 也随之增加，这些趋势与几位研究人员已经报告的趋势相似，因为高分辨率 SAR 图像似乎是这样的。显示结构寻找森林的图像纹理和森林信息之间的关系，结果发现图像幅度服从K分布的概率密度函数，并且K分布的阶参数之间存在很强的相关性。此外，从回归曲线来看，顺序参数随着生物量的增加而增加，达到 100 吨/公顷左右，远远超出了传统 RCS 方法的饱和极限。对未知森林的值进行估计并与 2005-2006 年地面测量值进行比较，得出模型准确度为 86%，最后使用所有地面测量数据更新回归模型。被认为对于估算北海道整个地区平地针叶林的生物量是有效的，并且通过将传统的 RCS 技术与纹理相结合，可以将森林生物量的估算精度提高到更高的水平；本研究中进行的分析较少。

项目成果

期刊论文数量（0）

专著数量（0）

科研奖励数量（0）

会议论文数量（0）

专利数量（0）

On the accuracy of empirical relation between forest biomass and order parameter of K-distribution in Pi-SAR images

论Pi-SAR图像中森林生物量与K分布序参数经验关系的准确性

DOI：
发表时间：
2006
期刊：
IEICE Tech. Rep. (WSANE 2006, X' an, China) 2006-21
影响因子：
0
作者：
H.Wang;et al.
通讯作者：
et al.

In Search of the Statistical Properties of High-Resolution Polarimetric SAR Data for the Measurements of Forest Biomass Beyond the RCS Saturation Limits

DOI：
10.1109/lgrs.2006.878299
发表时间：
2006-10
期刊：
IEEE Geoscience and Remote Sensing Letters
影响因子：
4.8
作者：
Haipeng Wang;K. Ouchi;Manabu Watanabe;M. Shimada;T. Tadono;A. Rosenqvist;S. Romshoo;M. Matsuoka;T. Moriyama;S. Uratsuka
通讯作者：
Haipeng Wang;K. Ouchi;Manabu Watanabe;M. Shimada;T. Tadono;A. Rosenqvist;S. Romshoo;M. Matsuoka;T. Moriyama;S. Uratsuka

On the accuracy of the empirical model for estimating forest biomass from K-distributed SAR images

K分布SAR影像估算森林生物量经验模型的准确性研究