Forest environment monitoring using texture analysis of high-resolution satellite data

利用高分辨率卫星数据纹理分析进行森林环境监测

• 批准号: 13660142
• 负责人: TSUYUKI Satoshi
• 金额: $ 2.24万
• 依托单位: The University of Tokyo
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (C)
• 财政年份: 2001
• 资助国家: 日本
• 起止时间: 2001 至 2002
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-13660142/
• 关键词: QuickBird; IKONOS; natural forest; segmentation; tree canopy density; mixed forest; 人工林; 立木本数; 樹種区分; 林相区分

It is considered very difficult to identify precise forest type classification of natural forest due to limitation of manual interpretation of aerial photograph or limitation of ground resolution of satellite imagery, though high-resolution satellite data will be the break through of such limitations. Permanent plots in the Tokyo University Forest in Hokkaido was employed as the test site of this study. IKONOS data acquired on August 16, 2001 and QuickBird data acquired on June 7, 2002 was used.In the first year, IKONOS data was used for analysis. The statistical value of digital number of all spectral bands within each tree crown was calculated to identify tree species. Tree species were classified into two groups of broad leave trees and needle leaf trees. According to statistical test, it is well separated between groups, but it was difficult with in each group.In the second year, QuickBird data was analyzed using textural segmentation method. It was found out that by segmentation method, individual tree crown was not recognized but several grouping level such as by log price or by tree family group could be identified.From this study, it was clarified that large scale mixed natural forest condition can be monitored using high-resolution remote sensing data.

由于航片人工判读的局限性或卫星图像地面分辨率的限制，天然林的精确林型划分一直被认为是非常困难的，而高分辨率卫星数据将是突破这些限制的突破口。在北海道的东京大学森林的永久样地作为本研究的试验地点。使用2001年8月16日获取的IKONOS数据和2002年6月7日获取的QuickBird数据，第一年使用IKONOS数据进行分析。计算每个树冠内所有光谱波段的数字数统计值来识别树种。树种可分为阔叶树和针叶树两大类。第二年对QuickBird数据进行了纹理分割分析。结果表明，分割方法不能识别出单个树冠，但可以识别出多个分组级别，如原木价格或家系组，从而阐明了利用高分辨率遥感数据可以监测大尺度天然混交林的状况。

项目成果

Sasakawa, H. and Tsuyuki, S.: "Development of forest function classification method using high-resolution remote sensing data (II)"Proceedings of 114th Japanese Forest Society Annual Conference. 427. (2003)

Sasakawa, H. 和 Tsuyuki, S.：“利用高分辨率遥感数据开发森林功能分类方法（II）”第 114 届日本林业协会年会论文集。

笹川裕史, 露木 聡: "高解像度衛星データを用いた森林機能区分手法の開発(II)-天然林林相区分手法の開発-"日本林学会学術講演集. 114. 427 (2003)

Hiroshi Sasakawa、Satoshi Tsuyuki：“利用高分辨率卫星数据开发森林功能分类方法（II） - 开发天然森林相分类方法”日本林业学会会议记录 114. 427 (2003)。

笹川裕史, 露木 聡: "高解像度衛星データを用いた森林機能区分手法の開発(II)-天然林林相区分手法の開発-"日本林学会論文集. 114(印刷中). (2003)

Hiroshi Sasakawa，Satoshi Tsuyuki：“利用高分辨率卫星数据开发森林功能分类方法（II） - 开发天然森林相分类方法 -”日本林业协会交易 114（出版中）。

田村 徹, 露木 聡, 廣川俊英, 平田雅和: "IKONOSデータによる林相区分手法の開発(I)東京大学北海道演習林における樹種区分"日本林学会学術講演集. 113. 320 (2002)

Toru Tamura、Satoshi Tsuyuki、Toshihide Hirokawa、Masakazu Hirata：“利用 IKONOS 数据开发森林相分类方法（I）东京大学北海道实验林的树种分类”日本林学会学术会议 113. 320。 (2002)

田村徹, 露木聡, 廣川俊英, 平田雅和: "IKONOSデータによる林相区分手法の開発(I) 東京大学北海道演習林における樹種区分"日本林学会論文集. 113(印刷中). (2002)

Toru Tamura、Satoshi Tsuyuki、Toshihide Hirokawa、Masakazu Hirata：“利用 IKONOS 数据开发森林相分类方法 (I) 东京大学北海道实验林中的树种分类” 日本林业学会汇刊 113（出版中） ）（2002）。