基于航空正射影像的面向对象林隙识别
|
摘要
【目的】研究分割尺度及对象特征对航空正射影像面向对象林隙识别的影响,评价基于航空正射影像林隙识别适宜性。【方法】以真彩色航空正射影像为基础数据,采用面向对象分类方法,以东北典型天然次生林帽儿山实验林场东林施业区为试验区进行林隙识别。在面向对象分类过程中,对航空正射影像的3个分量(Blue、Green和Red)采用10种尺度(10～100,步长为10)进行分割,应用拓扑组合准确度(R_(A(or))和R_(A(os)))和位置准确度(D_(sr))评价分割结果。对每种尺度分割结果,应用航空正射影像的光谱及光谱+纹理结合特征,采用带有线性核的支持向量机(SVM)分类器进行林隙、非林隙和树冠分类,共获得20种分类结果。利用混淆矩阵计算的生产者精度、用户精度、分类总精度和Kappa系数4个评价指标对分类结果进行精度评价。【结果】对象特征(大小和形状)受尺度参数影响,小尺度分割产生小面积对象,大尺度分割产生大面积对象,但大尺度不能有效将明显的林隙从冠层中分离出来,存在明显分割不足的现象。拓扑组合准确度(R_(A(or))和R_(A(os)))和位置准确度(D_(sr))说明与参考对象最相似的分割对象结果是在尺度20时获得的分割结果(R_(A(or))和R_(A(os))较大且接近,D_(sr)较小)。光谱及光谱+纹理结合特征分类方案分类总精度具有相似的变化特征,即小尺度分类总精度较低,随着尺度增大,分类总精度也再提高并在某个尺度达到最大值,之后分类总精度随尺度增大而降低,并趋于平稳。光谱+纹理结合特征的分类总精度低于仅使用光谱特征的分类总精度,在中小尺度上尤其明显。在尺度参数为20时使用光谱+纹理结合特征分类总精度低19个百分点,在尺度参数为30时低13个百分点。基于航空正射影像分割最优尺度参数为20。【结论】基于航空正射影像进行林隙识别,最高精度仅为74%(Kappa=61%),林隙生产者和用户精度在60%以上,非林隙生产者和用户精度在90%左右。基于航空正射影像林隙识别纹理特征的加入还将继续降低识别精度。
【Objective】 The effects of different segmentation scales and characteristics of objects on object-oriented forest gap based on aerial orthophoto were studied,and the suitability of forest gap classification based on aerial orthophoto was also evaluated in this research.【Method】 In this study, based on true color aerial orthophoto data and object-oriented classification method, northeastern typical natural secondary forest—Mao'ershan Experimental Forest Farm Donglin industry zone was selected as the study area for classification of forest gap. In the process of object-oriented classification, for three components of aerial orthophoto(Blue,Green,Red),10 scales(10-100, step size 10) was divided to carry out segmentation, and topological combination accuracy(R_(A(or)) and R_(A(os))) and position accuracy(D_(sr))were used to evaluate the classification results. For each segmentation result of different scales, spectral features and combination features of spectrum and texture derived from aerial orthophoto were used, and support vector machine(SVM) with linear kernel classifier of object-oriented was used to classify the study area into forest gap, non-forest gap and canopy, totally 20 classification result were obtained. Later the accuracy of each classification result was evaluated by the 4 different evaluation indexes, namely producer accuracy calculated by Confusion matrix, user accuracy, classification accuracy and Kappa coefficient.【Result】 Characteristics of objects(size and shape) were affected by scale parameter. Objects in small area were created by small scale segmentation, and objects in big area were created on large scale segmentation, however, segmentation of large scale could not efficiently distinguish obvious forest gaps from canopies, it indicates an obvious insufficient segmentation phenomenon. The topological combination accuracy(R_(A(or)) and R_(A(os)))and position accuracy(D_(sr))illustrated that the result of segmentation objects on 20 scale(R_(A(or)) and R_(A(os)) were higher and close, D_(sr) was lower) was most similar to the reference objects comparing to the result on other segmentation scales. The segmentation schemes of spectral features and combination features of spectrum and texture showed a similar variable trend: classification accuracies were lower on small scales, and then increased with scale increasing, after reached the maximum on a certain scale, they decreased to stable state. The classification accuracy using combination features of spectrum and texture was lower than that using spectral features only, especially on small scales. On the segmentation scale of 20, the overall classification accuracy using combination features of spectrum and texture was 19% lower than that using only spectral features, and accordingly 30% lower on the segmentation scale of 30. The optimal scale parameter of segmentation based on aerial orthophoto was 20.【Conclusion】 The highest recognition accuracy was just 74%(Kappa=0.61) when using aerial orthophoto to recognize forest gaps, the producer and user accuracy of forest gap were all above 60%, and which of non-forest gap were around 90%. The classification accuracy based on aerial orthophoto would continually decrease when adding texture features.
【Objective】 The effects of different segmentation scales and characteristics of objects on object-oriented forest gap based on aerial orthophoto were studied,and the suitability of forest gap classification based on aerial orthophoto was also evaluated in this research.【Method】 In this study, based on true color aerial orthophoto data and object-oriented classification method, northeastern typical natural secondary forest—Mao'ershan Experimental Forest Farm Donglin industry zone was selected as the study area for classification of forest gap. In the process of object-oriented classification, for three components of aerial orthophoto(Blue,Green,Red),10 scales(10-100, step size 10) was divided to carry out segmentation, and topological combination accuracy(R_(A(or)) and R_(A(os))) and position accuracy(D_(sr))were used to evaluate the classification results. For each segmentation result of different scales, spectral features and combination features of spectrum and texture derived from aerial orthophoto were used, and support vector machine(SVM) with linear kernel classifier of object-oriented was used to classify the study area into forest gap, non-forest gap and canopy, totally 20 classification result were obtained. Later the accuracy of each classification result was evaluated by the 4 different evaluation indexes, namely producer accuracy calculated by Confusion matrix, user accuracy, classification accuracy and Kappa coefficient.【Result】 Characteristics of objects(size and shape) were affected by scale parameter. Objects in small area were created by small scale segmentation, and objects in big area were created on large scale segmentation, however, segmentation of large scale could not efficiently distinguish obvious forest gaps from canopies, it indicates an obvious insufficient segmentation phenomenon. The topological combination accuracy(R_(A(or)) and R_(A(os)))and position accuracy(D_(sr))illustrated that the result of segmentation objects on 20 scale(R_(A(or)) and R_(A(os)) were higher and close, D_(sr) was lower) was most similar to the reference objects comparing to the result on other segmentation scales. The segmentation schemes of spectral features and combination features of spectrum and texture showed a similar variable trend: classification accuracies were lower on small scales, and then increased with scale increasing, after reached the maximum on a certain scale, they decreased to stable state. The classification accuracy using combination features of spectrum and texture was lower than that using spectral features only, especially on small scales. On the segmentation scale of 20, the overall classification accuracy using combination features of spectrum and texture was 19% lower than that using only spectral features, and accordingly 30% lower on the segmentation scale of 30. The optimal scale parameter of segmentation based on aerial orthophoto was 20.【Conclusion】 The highest recognition accuracy was just 74%(Kappa=0.61) when using aerial orthophoto to recognize forest gaps, the producer and user accuracy of forest gap were all above 60%, and which of non-forest gap were around 90%. The classification accuracy based on aerial orthophoto would continually decrease when adding texture features.
引文
郎璞玫. 2008. 航空像片冠幅与地面直径的线性混合模型. 林业科学,44(3):40-44(Lang P M. 2008. Linear mixed model of aerial photo crown width and ground diameter. Scientia Silvae Sinicae, 44(3):40-44.[in Chinese])
卢军, 李凤日, 张会儒, 等. 2011. 帽儿山天然次生林主要阔叶树种叶量分布模拟.林业科学,47(12):114-120.(Lu J, Li F R, Zhang H R, et al. 2011. Simulation of foliage distribution for major broad-leaved species in secondary forest in Mao’er Mountain.Scientia Silvae Sinicae, 47(12):114-120. [in Chinese])
毛学刚,侯吉宇,范文义.2017.林隙主被动遥感协同自动识别. 林业科学,53(11):94-103.(Mao X G, Hou J Y, Fan W Y, et al. 2017. Object-based automatic recognition for forest gaps using aerial image and LiDAR data. Scientia Silvae Sinicae, 53(11): 94-103. [in Chinese])
毛学刚, 杜子涵, 刘家倩, 等. 2018. 基于面向对象的QuickBird遥感影像林隙分割与分类. 应用生态学报, 29(1):44-52.(Mao X G, Du Z H, Liu J Q, et al. 2018. Object-oriented segmentation and classification of forest gap based on QuickBird remote sensing image.Chinese Journal of Applied Ecology, 29(1): 44-52. [in Chinese])
王晓杰, 范文义, 李明泽,等. 2013. 基于大比例尺航片的树种识别及单株定位方法. 东北林业大学学报, 41(5):152-157.(Wang X J, Fan W Y, Li M Z, et al. 2013. Tree species extraction and individual spatial location based on large-scale aerial photo. Journal of Northeast Forestry University, 41(5):152-157. [in Chinese])
赵峰, 庞勇, 李增元, 等. 2009. 机载激光雷达和航空数码影像单木树高提取. 林业科学, 45(10):81-87.(Zhao F, Pang Y, Li Z Y, et al. 2009. Extraction of individual tree height using a combination of aerial digital camera imagery and LiDAR. Scientia Silvae Sinicae, 45(10):81-87.[in Chinese])
张伐伐, 李卫忠, 卢柳叶,等. 2012. SVM多窗口纹理土地利用信息提取技术. 遥感学报, 16(1): 67-78.(Zhang F F, Li W Z, Lu L Y, et al. 2012. Technologies of extracting land utilization information based on SVM method with multi-window texture. Journal of Remote Sensing, 16(1): 67-78. [in Chinese])
郑淑丹,郑江华,石明辉,等. 2014. 基于分形和灰度共生矩阵纹理特征的种植型药用植物遥感分类. 遥感学报,18(4): 868-886.(Zheng S D, Zheng J H, Shi M H,et al. 2014. Classification of cultivated Chinese medicinal plants based on fractal theory and gray level co-occurrence matrix textures. Journal of Remote Sensing, 18(4): 868-886. [in Chinese])
Asner G P, Keller M, Pereira R, et al. 2004. Canopy damage and recovery after selective logging in Amazonia: field and satellite studies. Ecological Applications, 14(sp4): S280-S298.
Bonnet S, Gaulton R, Lehaire F, et al. 2015. Canopy gap mapping from airborne laser scanning: an assessment of the positional and geometrical accuracy. Remote Sensing, 7(9):11267-11294.
Benz U C, Hofmann P, Willhauck G, et al. 2004.Multi-resolution, object-oriented fuzzy analysis of remote sensing data for GIS-ready information. International Journal of Photogrammetry and Remote Sensing, 58(3/4):239-258.
Clark M L, Clark D B, Roberts D A. 2004. Small-footprint lidar estimation of sub-canopy elevation and tree height in a tropical rain forest landscape. Remote Sensing of Environment, 91(1):68-89.
Ecognition B. 2010. User Guide. Definiens Imaging GmbH. Munich: Definiens Imaging.
Garrity S R, Vierling L A, Smith A M S, et al. 2014. Automatic detection of shrub location, crown area, and cover using spatial wavelet analysis and aerial photography. Canadian Journal of Remote Sensing, 34(sup2):S376-S384.
He Y H, Franklin S E, Guo X L, et al. 2009. Narrow-linear and small-area forest disturbance detection and mapping from high spatial resolution imagery. Journal of Applied Remote Sensing, 3(1):2701-2712.
Immitzer M, Atzberger C, Koukal T.2012. Tree species classification with random forest using very high spatial resolution 8-band worldview-2 satellite data. Remote Sensing, 4(9):2661-2693.
Janssen L L F, Wei F J M. 1994.Accuracy assessment of satellite derived land-cover data: a review. Photogrammetric Engineering and Remote Sensing,60(4):419-426.
Jackson R G, Foody G M, Quine C P. 2000. Characterising windthrown gaps from fine spatial resolution remotely sensed data. Forest Ecology and Management, 135(9):253-260.
Johansen K, Arroyo L A, Phinn S, et al. 2010. Comparison of geo-object based and pixel-based change detection of riparian environments using high spatial resolution multi-spectral imagery. Photogrammetric Engineering and Remote Sensing, 76(2):123-136.
Ke Y, Quackenbush L J, Im J. 2010.Synergistic use of QuickBird multispectral imagery and LIDAR data for object-based forest species classification. Remote Sensing of Environment, 114(6):1141-1154.
Kim M, Madden M, Warner T. 2008. Estimation of optimal image object size for the segmentation of forest stands with multispectral IKONOS imagery. Object-Based Image Analysis. Berlin: Springer.
Lucas R, Bunting P, Paterson M, et al. 2008. Classification of Australian forest communities using aerial photography, CASI and HyMap data. Remote Sensing of Environment, 112(5):2088-2103.
Mao X G, Hou J Y. 2018. Object-based forest gaps classification using airborne LiDAR data. Journal of Forestry Research, (sp4):1-11.
Malahlela O, Cho M A, Mutanga O. 2014. Mapping canopy gaps in an indigenous subtropical coastal forest using high-resolution WorldView-2 data. International Journal of Remote Sensing, 35(17):6397-6417.
M?ller M, Lymburner L, Volk M. 2007.The comparison index: a tool for assessing the accuracy of image segmentation. International Journal of Applied Earth Observation & Geoinformation, 9(3):311-321.
Negrón-Juárez R I, Chambers J Q, Marra D M, et al. 2011. Detection of subpixel treefall gaps with Landsat imagery in Central Amazon forests. Remote Sensing of Environment, 115(12):3322-3328.
Petersen S L, Stringham T K, Laliberte A S. 2005.Classification of willow species using large-scale aerial photography. Rangeland Ecology and Management, 58(6):582-587.
Sun X Y, Du H Q, Han N, et al. 2014. Synergistic use of Landsat TM and SPOT5 imagery for object-based forest classification. Journal of Applied Remote Sensing,8(1): 083550-083564.
St-Onge B, Vepakomma U, Sénécal J F, et al. 2014. Canopy gap detection and analysis with airborne laser scanning. Forestry applications of airborne laser scanning. Berlin: Springer Netherlands.
Vapnik V N. 2000. The nature of statistical learning theory. Berlin: Springer Verlag.
Vepakomma U, St-Onge B, Kneeshaw D. 2011. Boreal forest height growth response to canopy gap Openings-An assessment with multi-temporal LiDAR data. Ecological Applications A Publication of the Ecological Society of America, 21, 99-121.
Wang L, Sousa W P, Gong P. 2004. Integration of object-based and pixel-based classification for mapping mangroves with IKONOS imagery. International Journal of Remote Sensing, 25(24):5655-5668.
Yang J, Jones T, Caspersen J, et al. 2015. Object-based canopy gap segmentation and classification: quantifying the pros and cons of integrating optical and lidar data. Remote Sensing, 7(12):15917-15932.
卢军, 李凤日, 张会儒, 等. 2011. 帽儿山天然次生林主要阔叶树种叶量分布模拟.林业科学,47(12):114-120.(Lu J, Li F R, Zhang H R, et al. 2011. Simulation of foliage distribution for major broad-leaved species in secondary forest in Mao’er Mountain.Scientia Silvae Sinicae, 47(12):114-120. [in Chinese])
毛学刚,侯吉宇,范文义.2017.林隙主被动遥感协同自动识别. 林业科学,53(11):94-103.(Mao X G, Hou J Y, Fan W Y, et al. 2017. Object-based automatic recognition for forest gaps using aerial image and LiDAR data. Scientia Silvae Sinicae, 53(11): 94-103. [in Chinese])
毛学刚, 杜子涵, 刘家倩, 等. 2018. 基于面向对象的QuickBird遥感影像林隙分割与分类. 应用生态学报, 29(1):44-52.(Mao X G, Du Z H, Liu J Q, et al. 2018. Object-oriented segmentation and classification of forest gap based on QuickBird remote sensing image.Chinese Journal of Applied Ecology, 29(1): 44-52. [in Chinese])
王晓杰, 范文义, 李明泽,等. 2013. 基于大比例尺航片的树种识别及单株定位方法. 东北林业大学学报, 41(5):152-157.(Wang X J, Fan W Y, Li M Z, et al. 2013. Tree species extraction and individual spatial location based on large-scale aerial photo. Journal of Northeast Forestry University, 41(5):152-157. [in Chinese])
赵峰, 庞勇, 李增元, 等. 2009. 机载激光雷达和航空数码影像单木树高提取. 林业科学, 45(10):81-87.(Zhao F, Pang Y, Li Z Y, et al. 2009. Extraction of individual tree height using a combination of aerial digital camera imagery and LiDAR. Scientia Silvae Sinicae, 45(10):81-87.[in Chinese])
张伐伐, 李卫忠, 卢柳叶,等. 2012. SVM多窗口纹理土地利用信息提取技术. 遥感学报, 16(1): 67-78.(Zhang F F, Li W Z, Lu L Y, et al. 2012. Technologies of extracting land utilization information based on SVM method with multi-window texture. Journal of Remote Sensing, 16(1): 67-78. [in Chinese])
郑淑丹,郑江华,石明辉,等. 2014. 基于分形和灰度共生矩阵纹理特征的种植型药用植物遥感分类. 遥感学报,18(4): 868-886.(Zheng S D, Zheng J H, Shi M H,et al. 2014. Classification of cultivated Chinese medicinal plants based on fractal theory and gray level co-occurrence matrix textures. Journal of Remote Sensing, 18(4): 868-886. [in Chinese])
Asner G P, Keller M, Pereira R, et al. 2004. Canopy damage and recovery after selective logging in Amazonia: field and satellite studies. Ecological Applications, 14(sp4): S280-S298.
Bonnet S, Gaulton R, Lehaire F, et al. 2015. Canopy gap mapping from airborne laser scanning: an assessment of the positional and geometrical accuracy. Remote Sensing, 7(9):11267-11294.
Benz U C, Hofmann P, Willhauck G, et al. 2004.Multi-resolution, object-oriented fuzzy analysis of remote sensing data for GIS-ready information. International Journal of Photogrammetry and Remote Sensing, 58(3/4):239-258.
Clark M L, Clark D B, Roberts D A. 2004. Small-footprint lidar estimation of sub-canopy elevation and tree height in a tropical rain forest landscape. Remote Sensing of Environment, 91(1):68-89.
Ecognition B. 2010. User Guide. Definiens Imaging GmbH. Munich: Definiens Imaging.
Garrity S R, Vierling L A, Smith A M S, et al. 2014. Automatic detection of shrub location, crown area, and cover using spatial wavelet analysis and aerial photography. Canadian Journal of Remote Sensing, 34(sup2):S376-S384.
He Y H, Franklin S E, Guo X L, et al. 2009. Narrow-linear and small-area forest disturbance detection and mapping from high spatial resolution imagery. Journal of Applied Remote Sensing, 3(1):2701-2712.
Immitzer M, Atzberger C, Koukal T.2012. Tree species classification with random forest using very high spatial resolution 8-band worldview-2 satellite data. Remote Sensing, 4(9):2661-2693.
Janssen L L F, Wei F J M. 1994.Accuracy assessment of satellite derived land-cover data: a review. Photogrammetric Engineering and Remote Sensing,60(4):419-426.
Jackson R G, Foody G M, Quine C P. 2000. Characterising windthrown gaps from fine spatial resolution remotely sensed data. Forest Ecology and Management, 135(9):253-260.
Johansen K, Arroyo L A, Phinn S, et al. 2010. Comparison of geo-object based and pixel-based change detection of riparian environments using high spatial resolution multi-spectral imagery. Photogrammetric Engineering and Remote Sensing, 76(2):123-136.
Ke Y, Quackenbush L J, Im J. 2010.Synergistic use of QuickBird multispectral imagery and LIDAR data for object-based forest species classification. Remote Sensing of Environment, 114(6):1141-1154.
Kim M, Madden M, Warner T. 2008. Estimation of optimal image object size for the segmentation of forest stands with multispectral IKONOS imagery. Object-Based Image Analysis. Berlin: Springer.
Lucas R, Bunting P, Paterson M, et al. 2008. Classification of Australian forest communities using aerial photography, CASI and HyMap data. Remote Sensing of Environment, 112(5):2088-2103.
Mao X G, Hou J Y. 2018. Object-based forest gaps classification using airborne LiDAR data. Journal of Forestry Research, (sp4):1-11.
Malahlela O, Cho M A, Mutanga O. 2014. Mapping canopy gaps in an indigenous subtropical coastal forest using high-resolution WorldView-2 data. International Journal of Remote Sensing, 35(17):6397-6417.
M?ller M, Lymburner L, Volk M. 2007.The comparison index: a tool for assessing the accuracy of image segmentation. International Journal of Applied Earth Observation & Geoinformation, 9(3):311-321.
Negrón-Juárez R I, Chambers J Q, Marra D M, et al. 2011. Detection of subpixel treefall gaps with Landsat imagery in Central Amazon forests. Remote Sensing of Environment, 115(12):3322-3328.
Petersen S L, Stringham T K, Laliberte A S. 2005.Classification of willow species using large-scale aerial photography. Rangeland Ecology and Management, 58(6):582-587.
Sun X Y, Du H Q, Han N, et al. 2014. Synergistic use of Landsat TM and SPOT5 imagery for object-based forest classification. Journal of Applied Remote Sensing,8(1): 083550-083564.
St-Onge B, Vepakomma U, Sénécal J F, et al. 2014. Canopy gap detection and analysis with airborne laser scanning. Forestry applications of airborne laser scanning. Berlin: Springer Netherlands.
Vapnik V N. 2000. The nature of statistical learning theory. Berlin: Springer Verlag.
Vepakomma U, St-Onge B, Kneeshaw D. 2011. Boreal forest height growth response to canopy gap Openings-An assessment with multi-temporal LiDAR data. Ecological Applications A Publication of the Ecological Society of America, 21, 99-121.
Wang L, Sousa W P, Gong P. 2004. Integration of object-based and pixel-based classification for mapping mangroves with IKONOS imagery. International Journal of Remote Sensing, 25(24):5655-5668.
Yang J, Jones T, Caspersen J, et al. 2015. Object-based canopy gap segmentation and classification: quantifying the pros and cons of integrating optical and lidar data. Remote Sensing, 7(12):15917-15932.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |