Search

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.

The potential of very high spatial resolution Pléiades image texture for forest structure quantification and mapping was assessed on maritime pine stands in south-western France. A preliminary step showed that multi-linear regressions allowed a reliable prediction of forest variables (such as crown diameter or tree height) from a set of features automatically selected among a huge number of texture features with various spatial parameterizations. The image classification was performed using the Random Forests (RF) ensemble classifier in order to discriminate five forest structure classes using a hierarchical approach. The RF-variable importance is used for texture feature selection. The results highlight the contribution of process automation and the need for a joint use of both Pléiades image resolutions (panchromatic and multispectral) to derive the best performing texture features., Cette étude montre le potentiel de l'information texturale des images à très haute résolution spatiale Pléiades pour la quantification et la cartographie de la structure forestière des peuplements de pin maritime du sud-ouest de la France (massif forestier landais). Une première étape montre qu'il est possible d'estimer, par régressions linéaires multiples, les variables de structure forestière (comme le diamètre des couronnes ou la hauteur des arbres) à partir d'un ensemble d'attributs de texture automatiquement sélectionnés parmi un grand nombre de paramétrages possibles. La classification de l'image est ensuite effectuée en utilisant l'algorithme des forêts aléatoires (RF) pour discriminer cinq classes de structure forestière avec une approche hiérarchique. L'importance de variable des RF est utilisée pour la sélection des attributs de texture. Les résultats montrent l'intérêt de l'automatisation du processus et de l'utilisation conjointe des deux résolutions des images Pléiades (mode panchromatique et mode multispectral) pour dériver les attributs de texture les plus performants pour détecter de fines variations de structure forestière.

International audience; Since several years, more and more studies aim at developing phenology products from satellite time-series at high temporal frequency such as those provided by the VEGETATION or MODIS sensors. Reflectance times-series at high spatial resolution, such as those that will be obtained from SENTINEL2, will soon available to monitor the phenology response of forests under climate change at the level of the forest stand or the tree species. There is a great need for continuous in situ monitoring of phenology to calibrate and validate the current and future remotely-sensed phenology products. In this context, we proposed a method based on near surface remote sensing techniques to monitor the seasonal change in LAI (Leaf area Index) and date key stages of the leaf phenology. As it is important to use a network of sensors with autonomous recording systems at a minimal cost in order to maximize the spatial sampling, we selected a method based on the transmittance continuous measurement of photosynthetic active radiation (PAR). We evaluated and validated the method for a deciduous forest species (Quercus petraea) over two sites encompassing a large variation in the timing of spring leafing, where direct visual phenology observations were performed. A specific preprocessing and modeling of the measured temporal signal was developed. The performances for dating the leaf unfolding in spring are satisfying: the bias is lower than ∼1 day and the RMSE is (generally) lower than ∼ 4 days.

This paper presents a new feature selection method which aims to effectively and efficiently map remote sensing data. An automated texture-based modelling procedure of forest structure variables is at the core of our approach. We show that texture features that are highly correlated to genuine physical parameters of forest structure have potential for building reliable classifiers. We demonstrate the effectiveness of our modelling-based texture feature selection method in performing mapping of very high resolution forest images. Our method outperforms Random Forest variable importance in terms of classification accuracy and computational complexity.

International audience; Over the last decade, a growing number of Very High Resolution (VHR) remote sensing data from various spatial sensors has become available. They provide interesting information for forest inventory applications thanks to the strong relationship between forest spatial structure and image texture at stand level. The main goal of this study is to find the most relevant image features to describe forest structure from very high resolution satellite imagery. The emphasis is placed on the automatisation of this process, exploiting both spectral and spatial information. Compared to High Resolution images, VHR imagery provides a very rich textural information that has to be thoroughly investigated for the forest structure characterization. Our approach is based on linear regressions between the forest structure variables to be estimated and various spectral and Haralick's texture features (derived from Grey Level Co-occurrence Matrix). The main drawback of this well-known texture representation is the underlying parameters (window size, displacement length and orientation, quantification level) which are extremely difficult to set due to the spatial complexity of the forest structure. To tackle this major issue, probably the main cause of poor texture analysis in practice, we propose an automatic feature selection process whose originality lies on the use of test frames of adequate forest samples where the structure variables were measured from ground. This method, inspired by camera calibration protocols, selects the best image descriptors via statistical modelling, exploring a wide range of parameter values. Hence, just a few samples are required to build up the test frames but allow a fast assessment of thousands of descriptors, given the large number of tested combinations of parameters values. This method has been successfully applied to the modelling of 7 typical forest structure variables (age, tree height, crown diameter, diameter at breast hight, basal area, density and tree spacing) over maritime pine stands in South West of France from QuickBird panchromatic and multi-spectral images. Overall results are good, multi-spectral and panchromatic images show similar performances to provide well-suited features to estimate the forest variables. The coefficient of determination, R², of the best models for 6 of the forest variables of interest, estimated from the test frames, ranges from 0.89 to 0.97. Only the basal area was weakly correlated to the considered image features (0.64). To improve the results, various linear combinations of panchromatic descriptors or multi-spectral descriptors or a combination of both, were tested using multiple linear regressions. As collinearity is a very perturbing problem in multi-linear regression, this issue is carefully addressed. Different variables subset selection methods are tested. A stepwise method, derived from LARS (Least Angular Regression), turned out the most convincing, significantly improving the quality of estimation for all the forest structure variables, including the basal area (R²>0.98). The best estimation results are obtained from subsets combining multi-spectral and panchromatic features, highlighting the potential of a multi-scale approach for retrieving forest structure variables from VHR satellite images.