Automated retrieval of forest structure variables based on multi-scale texture analysis of VHR satellite imagery

Abstract The main goal of this study is to design a method to describe the structure of forest stands from Very High Resolution satellite imagery, relying on some typical variables such as crown diameter, tree height, trunk diameter, tree density and tree spacing. The emphasis is placed on the automatization of the process of identification of the most relevant image features for the forest structure retrieval task, exploiting both spectral and spatial information. Our approach is based on linear regressions between the forest structure variables to be estimated and various spectral and Haralick’s texture features. The main drawback of this well-known texture representation is the underlying parameters which are extremely difficult to set due to the spatial complexity of the forest structure. To tackle this major issue, an automated feature selection process is proposed which is based on statistical modeling, exploring a wide range of parameter values. It provides texture measures of diverse spatial parameters hence implicitly inducing a multi-scale texture analysis. A new feature selection technique, we called Random PRiF, is proposed. It relies on random sampling in feature space, carefully addresses the multicollinearity issue in multiple-linear regression while ensuring accurate prediction of forest variables. Our automated forest variable estimation scheme was tested on Quickbird and Pleiades panchromatic and multispectral images, acquired at different periods on the maritime pine stands of two sites in South-Western France. It outperforms two well-established variable subset selection techniques. It has been successfully applied to identify the best texture features in modeling the five considered forest structure variables. The RMSE of all predicted forest variables is improved by combining multispectral and panchromatic texture features, with various parameterizations, highlighting the potential of a multi-resolution approach for retrieving forest structure variables from VHR satellite images. Thus an average prediction error of ∼ 1.1 m is expected on crown diameter, ∼ 0.9 m on tree spacing, ∼ 3 m on height and ∼ 0.06 m on diameter at breast height.

[1]  N. Chehata,et al.  OBJECT-BASED FOREST CHANGE DETECTION USING HIGH RESOLUTION SATELLITE IMAGES , 2013 .

[2]  Conghe Song,et al.  Estimating average tree crown size using spatial information from Ikonos and QuickBird images: Across-sensor and across-site comparisons , 2010 .

[3]  Frieke Van Coillie,et al.  Feature selection by genetic algorithms in object-based classification of IKONOS imagery for forest mapping in Flanders, Belgium , 2007 .

[4]  Volker C. Radeloff,et al.  The Impact of Phenological Variation on Texture Measures of Remotely Sensed Imagery , 2009, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[5]  J. Hyyppä,et al.  Accuracy comparison of various remote sensing data sources in the retrieval of forest stand attributes , 2000 .

[6]  Samia Boukir,et al.  RETRIEVING FOREST STRUCTURE VARIABLES FROM VERY HIGH RESOLUTION SATELLITE IMAGES USING AN AUTOMATIC METHOD , 2012 .

[7]  Ulrike Groemping,et al.  Relative Importance for Linear Regression in R: The Package relaimpo , 2006 .

[8]  Jean Louchet,et al.  Using colour, texture, and hierarchial segmentation for high-resolution remote sensing , 2008 .

[9]  M. Castillo-Santiago,et al.  Estimation of tropical forest structure from SPOT-5 satellite images , 2010 .

[10]  Paolo Gamba,et al.  Texture-based characterization of urban environments on satellite SAR images , 2003, IEEE Trans. Geosci. Remote. Sens..

[11]  C. Proisy,et al.  Predicting and mapping mangrove biomass from canopy grain analysis using Fourier-based textural ordination of IKONOS images , 2007 .

[12]  David M. Allen,et al.  The Relationship Between Variable Selection and Data Agumentation and a Method for Prediction , 1974 .

[13]  Mohan M. Trivedi,et al.  Segmentation of a high-resolution urban scene using texture operators , 1984, Comput. Vis. Graph. Image Process..

[14]  B. Manly.  Multivariate Statistical Methods : A Primer , 1986 .

[15]  V. Radeloff,et al.  Image texture as a remotely sensed measure of vegetation structure , 2012 .

[16]  L. Ruiz,et al.  TEXTURE FEATURE EXTRACTION FOR CLASSIFICATION OF REMOTE SENSING DATA USING WAVELET DECOMPOSITION : A COMPARATIVE STUDY , 2004 .

[17]  D. Guyon,et al.  Estimation de caractéristiques forestières à partir d'images à haute résolution spatiale (SPOT 5) , 1996 .

[18]  Ryan R. Jensen,et al.  Using remote sensing image texture to study habitat use patterns: a case study using the polymorphic white‐throated sparrow (Zonotrichia albicollis) , 2006 .

[19]  S. Franklin,et al.  Regenerating boreal forest structure estimation using SPOT‐5 pan‐sharpened imagery , 2007 .

[20]  P. Maillard.  Comparing Texture Analysis Methods through Classification , 2003 .

[21]  Brian R. Sturtevant,et al.  Estimation of forest structural parameters using 5 and 10 meter SPOT-5 satellite data , 2009 .

[22]  Jean-Pierre Wigneron,et al.  Détection de changements structurels sur des images satellite haute r ésolution. Application en milieu forestier , 2013, Traitement du Signal.

[23]  J. Neter,et al.  Applied Linear Regression Models , 1983 .

[24]  Wilhelm Burger,et al.  Digital Image Processing - An Algorithmic Introduction using Java , 2008, Texts in Computer Science.

[25]  Sakari Tuominen,et al.  Performance of different spectral and textural aerial photograph features in multi-source forest inventory , 2005 .

[26]  Leo R Beard,et al.  Statistical Methods in Hydrology , 1962 .

[27]  William J. Emery,et al.  A neural network approach using multi-scale textural metrics from very high-resolution panchromatic imagery for urban land-use classification , 2009 .

[28]  B. St-Onge.  Automated forest structure mapping from high resolution imagery based on directional semivariogram estimates , 1997 .

[29]  Giuseppe Scarpa,et al.  Hierarchical Texture-Based Segmentation of Multiresolution Remote-Sensing Images , 2009, IEEE Transactions on Geoscience and Remote Sensing.

[30]  Zeng-yuan Li,et al.  Estimation of stand mean crown diameter from high-spatial-resolution imagery based on a geostatistical method , 2010 .

[31]  P. Couteron,et al.  Predicting tropical forest stand structure parameters from Fourier transform of very high‐resolution remotely sensed canopy images , 2005 .

[32]  S. Franklin.  Using spatial Co-occurrence texture to increase forest structure and species composition classification accuracy , 2001 .

[33]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[34]  L. Leemis.  Applied Linear Regression Models , 1991 .

[35]  C. Proisy,et al.  The variation of apparent crown size and canopy heterogeneity across lowland Amazonian forests , 2010 .

[36]  Denis Loustau,et al.  Estimating the foliage area of maritime pine (Pinus pinaster Aït.) branches and crowns with application to modelling the foliage area distribution in the crown. , 2000 .

[37]  Samia Boukir,et al.  Modelling-Based Feature Selection for Classification of Forest Structure Using Very High Resolution Multispectral Imagery , 2013, 2013 IEEE International Conference on Systems, Man, and Cybernetics.

[38]  Cristina Gómez,et al.  Modeling Forest Structural Parameters in the Mediterranean Pines of Central Spain using QuickBird-2 Imagery and Classification and Regression Tree Analysis (CART) , 2012, Remote. Sens..

[39]  Arnon Karnieli,et al.  redicting forest structural parameters using the image texture derived from orldView-2 multispectral imagery in a dryland forest , Israel , 2011 .

[40]  S. Franklin,et al.  Aerial Image Texture Information in the Estimation of Northern Deciduous and Mixed Wood Forest Leaf Area Index (LAI) , 1998 .

[41]  Craig A. Coburn,et al.  A multiscale texture analysis procedure for improved forest stand classification , 2004 .

[42]  N. Chehataa,et al.  Object-based change detection in wind storm-damaged forest using high-resolution multispectral images , 2014 .

[43]  R. Tibshirani,et al.  Least angle regression , 2004, math/0406456.

[44]  N. Lam,et al.  Wavelets for Urban Spatial Feature Discrimination: Comparisons with Fractal, Spatial Autocorrelation, and Spatial Co-Occurrence Approaches , 2004 .

[45]  P. Defourny,et al.  Retrieving forest structure variables based on image texture analysis and IKONOS-2 imagery , 2006 .

[46]  Jean-Pierre Da Costa,et al.  Texture based image retrieval and classification of very high resolution maritime pine forest images , 2013, 2013 IEEE International Geoscience and Remote Sensing Symposium - IGARSS.

[47]  Arko Lucieer,et al.  Texture-based classification of sub-Antarctic vegetation communities on Heard Island , 2010, Int. J. Appl. Earth Obs. Geoinformation.