加藤 顕

High resolution active remote sensing data are getting more accessible. ALOS PALSAR radar data can provide 10 to 30 m resolution data. With those resolutions, land use changes can be detected and monitored well. The land use map of multi-temporal data is made by radar images, but seasonal change caused by deciduousness should be separately identified from the human activities. In this study, airborne lidar is used as ground truth data to estimate canopy volume and the estimated volume is compared with the change on the backscattering coefficients of multi-temporal radar images. The process used in this study is useful to monitor the small change detection with high resolution radar data given by ALOS 2 satellite launch near the future.

千葉県の 5 つの地域においてモウソウチク林群落の分布を調査し，11 項目のモウソウチク林の群落や拡大箇所の特徴と侵略度の関係を検討した。その結果，侵略性への影響度の最も高い要因は拡大箇所の植生で，二番目に拡大箇所の傾斜方位であった。拡大箇所の植生では，ススキ－アズマネザサ群集への侵略度が高く，樹林の中では落葉広葉樹林への侵略度が高いことが分かった。また，竹林の管理状況と侵略性に関係は見られなかった。侵略性への影響度の高い 5 つの要因から数量化II類によって拡大モデル式を作成した結果，侵略性の強い群落を 68.6％の確率で判別することができた。

近年, 車載レーザーセンサーが利用可能となってきたが, 車載レーザーを用いた研究はあまり行われていない。本研究は高速道路法面を対象地として, 車載レーザーデータから樹木地への反射特性を把握するためにシミュレーション開発を行った。これまでの研究では, バイオマス量等をレーザーデータから推定する場合, 高さ別レーザー反射割合を求めて, パーセンタイルでの高さを説明変数とし, モデル化を行ってきた。そのパーセンタイルでの高さが照射視野によってどのように変化するかはわかっていない。そこで本研究では, 異なる照射視野からの照射シミュレーションを行い, 照射視野の違いによるパーセンタイルの高さを比較した。現地で測定したデータを基にコンピュータ上で同じ樹冠形状を再現し, 3 つのセンサー視野（上空からの照射, 側面照射, 固定点照射）から照射シミュレーションを行った。シミュレーションではレイ・トレーシング アルゴリズムを用い, 各照射視野からコンピュータ上で再現した樹木にバーチャルでレーザー照射を行い, レーザーが当たった樹冠上の場所を記録した。シミュレーション結果を検証するため 10 m×10 m のプロットを対象にシミュレーション結果と実測した車載レーザーデータとを比較した。側面照射によるシミュレーション結果が実測のデータと相関関係が高かったため, シミュレーションが有効であることがわかった。今後様々な傾斜角度の法面でデータを取得し, バイオマス計測を推定するための照射視野と傾斜角度の関係も明らかにしたい。

ALOS PALSAR uses L band which is related with volume scattering. The volume scattering is the reflection comes from the surface and the inside of canopy. Forest management activities can be monitored using PALSAR L-band for Reducing Emissions from Deforestation and Forest Degradation (REDD). But it is still unknown how the biomass changed by the thinning (forest management) influences the backscattering coefficients. In this study, the change of stem volume (biomass) was computed and mapped using multi-temporal airborne lidar data in wide area. The stem volume using field measured tree parameters were interpolated and mapped by the airborne lidar data and the mapped biomass was used as the ground truth for radar image to see the polarization change caused by the thinning. The polarization axis of HV/HH backscattering coefficients is shifted in 9.8 degrees. This axis change is a useful indicator for the end member analysis using HV/HH polarization.

Data fusion between aerial photos and LiDAR provides better estimates in forestry and ecological applications, because LiDAR mainly provides the structural information of objects and aerial photo can add spectral information to them. Without the data fusion, an accurate identification of tree crown information from two dimensional data is difficult due to shaded and shadow pixels cast on the image and image distortion. The aerial photogrammetric techniques cannot reconstruct the objects accurately in three dimensional spaces if they are not clearly visible on the photos. The conventional orthophotos, therefore, still have image distortion due to an inappropriate Digital Surface Model (DSM). LiDAR provides a more suitable surface of tree crown structure in three-imensional spaces. This LiDAR-derived DSM could be used in conjunction with conventional photogrammetric techniques to rectify aerial photos and produce true orthophotos for each image. The existence of different perspective points from the use of multiple images results in different illumination and shadows cast on the DSM from the angle between the sun and the camera. In previous studies, a Z-buffer algorithm was applied for the occlusion detection and compensation. However, the technique was computationally intensive. In this study, the camera view and sun-oriented hillshade were generated using the LiDAR-derived DSM. The hillshade surfaces distinguished between the exposed and the occluded side of the DSM during the composition process of respective true orthophotos. This technique constituted a simpler approach and is applicable to data fusion between LiDAR and multispectral imagery to make an orthographically rectified image.

Forest structure data derived from lidar is being used in forest science and management for inventory analysis, biomass estimation, and wildlife habitat analysis. Regression analysis dominated previous approaches to the derivation of tree stem and crown parameters from lidar. The regression model for tree parameters is locally applied based on vertical lidar point density the tree species involved, and stand structure in the specific research area. The results of this approach, therefore, are location-specific, limiting its applicability to other areas. For a more widely applicable approach to derive tree parameters, we developed an innovative method called 'wrapped surface reconstruction' that employs radial basis functions and an isosurface. Utilizing computer graphics, we capture the exact shape of an irregular tree crown of various tree species based on the lidar point cloud and visualize their exact crown formation in three-dimensional space. To validate the tree parameters given by our wrapped surface approach, survey-grade equipment (a total station) was used to measure the crown shape. Four vantage points were established for each of 55 trees to capture whole-tree crown profiles georeferenced with post-processed differential GPS points. The observed tree profiles were linearly interpolated to estimate crown volume. These fieldwork-generated profiles were compared with the wrapped surface to assess goodness of fit For coniferous trees, the following tree crown parameters derived by the wrapped surface method were highly correlated (p<0.05) with the total station-derived measurements: tree height (R(2)=0.95), crown width (R(2)=0.80), live crown base (R(2)=0.92), height of the lowest branch (R(2)=0.72), and crown volume (R(2)=0.84). For deciduous trees, wrapped surface-derived parameters of tree height (R(2)=0.96), crown width (R(2)=0.75), live crown base (R(2)=0.53), height of the lowest branch (R(2)=0.51), and crown volume (R(2)=0.89) were correlated with the total station-derived measurements. The wrapped surface technique is less susceptible to errors in estimation of tree parameters because of exact interpolation using the radial basis functions. The effect of diminished energy return causes the low correlation for lowest branches in deciduous trees (R(2)=0.51), even though leaf-off lidar data was used. The wrapped surface provides fast and automated detection of micro-scale tree parameters for specific applications in areas such as tree physiology, fire modeling, and forest inventory. (C) 2009 Elsevier Inc. All rights reserved.