Veraldo Liesenberg

Object-oriented and pixel-based classification approaches to classify tropical successional stages using airborne high–spatial resolution images

Airborne high–spatial resolution images were evaluated for mapping purposes in a complex Atlantic rainforest environment in southern Brazil. Two study sites, covered predominantly by secondary evergreen rainforest, were surveyed by airborne multispectral high-resolution imagery. These aerophotogrammetric images were acquired at four spectral bands (visible to near-infrared) with spatial resolution of 0.39 m. We evaluated different data input scenarios to suit the object-oriented classification approach. In addition to the four spectral bands, auxiliary products such as band ratios and digital elevation models were considered. Comparisons with traditional pixel-based classifiers were also performed. The results showed that the object-based classification approach yielded a better overall accuracy, ranging from 89% to 91%, than the pixel-based classifications, which ranged from 62% to 63%. The individual classification accuracy of forest-related classes, such as young successional forest stages, benefits the object-based approach. These classes have been reported in the literature as the most difficult to map in tropical environments. The results confirm the potential of object-based classification for mapping procedures and discrimination of successional forest stages and other related land use and land cover classes in complex Atlantic rainforest environments. The methodology is suggested for further SAAPI acquisitions in order to monitor such endangered environment as well as to support National Land and Environmental Management Protocols.

Evaluating Sentinel-2 and Landsat-8 Data to Map Sucessional Forest Stages in a Subtropical Forest in Southern Brazil

Studies designed to discriminate different successional forest stages play a strategic role in forest management, forest policy and environmental conservation in tropical environments. The discrimination of different successional forest stages is still a challenge due to the spectral similarity among the concerned classes. Considering this, the objective of this paper was to investigate the performance of Sentinel-2 and Landsat-8 data for discriminating different successional forest stages of a patch located in a subtropical portion of the Atlantic Rain Forest in Southern Brazil with the aid of two machine learning algorithms and relying on the use of spectral reflectance data selected over two seasons and attributes thereof derived. Random Forest (RF) and Support Vector Machine (SVM) were used as classifiers with different subsets of predictor variables (multitemporal spectral reflectance, textural metrics and vegetation indices). All the experiments reached satisfactory results, with Kappa indices varying between 0.9, with Landsat-8 spectral reflectance alone and the SVM algorithm, and 0.98, with Sentinel-2 spectral reflectance alone also associated with the SVM algorithm. The Landsat-8 data had a significant increase in accuracy with the inclusion of other predictor variables in the classification process besides the pure spectral reflectance bands. The classification methods SVM and RF had similar performances in general. As to the RF method, the texture mean of the red-edge and SWIR bands were considered the most important ranked attributes for the classification of Sentinel-2 data, while attributes resulting from multitemporal bands, textural metrics of SWIR bands and vegetation indices were the most important ones in the Landsat-8 data classification.

Análise comparativa de classificadores digitais em imagens do Landsat-8 aplicados ao mapeamento temático

The objective of this work was to evaluate the performance of SVM and K‑NN digital classifiers for the object‑based classification on Landsat‑8 images, applied to mapping of land use and land cover of Alta Bacia do Rio Piracicaba‑Jaguari, in the state of Minas Gerais, Brazil. The pre‑processing step consisted of using radiometric conversion and atmospheric correction. Then the multispectral bands (30 m) were merged with the panchromatic band (15 m). Based on RGP compositions and field inspection, 15 land‑use and land‑cover classes were defined. For edge segmentation, the bounds were set to 10 and 60 for segmentation configuring and merging in the ENVI software. Classification was done using SVM and K‑NN. Both classifiers showed high values for the Kappa index (k): 0.92 for SVM and 0.86 for K‑NN, significantly different from each other at 95% probability. A major improvement was observed for SVM by the correct classification of different forest types. The object‑based classification is largely applied on high‑resolution spatial images; however, the results of the present work show the robustness of the method also for medium‑resolution spatial images.

Multifrequency and Full-Polarimetric SAR Assessment for Estimating Above Ground Biomass and Leaf Area Index in the Amazon Várzea Wetlands

The aim of this study is to evaluate the potential of multifrequency and Full-polarimetric Synthetic Aperture Radar (SAR) data for retrieving both Above Ground Biomass (AGB) and Leaf Area Index (LAI) in the Amazon floodplain forest environment. Two specific questions were proposed: (a) Does multifrequency SAR data perform more efficiently than single-frequency data in estimating LAI and AGB of várzea forests?; and (b) Are quad-pol SAR data more efficient than singleand dual-pol SAR data in estimating LAI and AGB of várzea forest? To answer these questions, data from different sources (TerraSAR-X Multi Look Ground Range Detected (MGD), Radarsat-2 Standard Qual-Pol, advanced land observing satellite (ALOS)/ phased-arrayed L-band SAR (PALSAR-1). Fine-beam dual (FDB) and quad Polarimetric mode) were combined in 10 different scenarios to model both LAI and AGB. A R-platform routine was implemented to automatize the selection of the best regression models. Results indicated that ALOS/PALSAR variables provided the best estimates for both LAI and AGB. Single-frequency L-band data was more efficient than multifrequency SAR. PALSAR-FDB HV-dB provided the best LAI estimates during low-water season. The best AGB estimates at high-water season were obtained by PALSAR-1 quad-polarimetric data. The top three features for estimating AGB were proportion of volumetric scattering and both the first and second dominant phase difference between trihedral and dihedral scattering, extracted from Van Zyl and Touzi decomposition, respectively. The models selected for both AGB and LAI were parsimonious. The Root Mean Squared Error (RMSEcv), relative overall RMSEcv (%) and R2 value for LAI were 0.61%, 0.55% and 13%, respectively, and for AGB, they were 74.6 t·ha−1, 0.88% and 46%, respectively. These results indicate that L-band (ALOS/PALSAR-1) has a high potential to provide quantitative and spatial information about structural forest attributes in floodplain forest environments. This potential may be extended not only with PALSAR-2 data but also to forthcoming missions (e.g., NISAR, Global Ecosystems Dynamics Investigation Lidar (GEDI), BIOMASS, Tandem-L) for promoting wall-to-wall AGB mapping with a high level of accuracy in dense tropical forest regions worldwide.

Comparing the Performance of Ground Filtering Algorithms for Terrain Modeling in a Forest Environment Using Airborne LiDAR Data

The aim of this study was to evaluate the performance of four ground filtering algorithms to generate digital terrain models (DTMs) from airborne light detection and ranging (LiDAR) data. The study area is a forest environment located in Washington state, USA with distinct classes of land use and land cover (e.g., shrubland, grassland, bare soil, and three forest types according to tree density and silvicultural interventions: closed-canopy forest, intermediate-canopy forest, and open-canopy forest). The following four ground filtering algorithms were assessed: Weighted Linear Least Squares (WLS), Multi-scale Curvature Classification (MCC), Progressive Morphological Filter (PMF), and Progressive Triangulated Irregular Network (PTIN). The four algorithms performed well across the land cover, with the PMF yielding the least number of points classified as ground. Statistical differences between the pairs of DTMs were small, except for the PMF due to the highest errors. Because the forestry sector requires constant updating of topographical maps, open-source ground filtering algorithms, such as WLS and MCC, performed very well on planted forest environments. However, the performance of such filters should also be evaluated over complex native forest environments.

Estimating Stand Height and Tree Density in Pinus taeda plantations using in-situ data, airborne LiDAR and k-Nearest Neighbor Imputation

Accurate forest inventory is of great economic importance to optimize the entire supply chain management in pulp and paper companies. The aim of this study was to estimate stand dominate and mean heights (HD and HM) and tree density (TD) of Pinus taeda plantations located in South Brazil using in-situ measurements, airborne Light Detection and Ranging (LiDAR) data and the non- k-nearest neighbor (k-NN) imputation. Forest inventory attributes and LiDAR derived metrics were calculated at 53 regular sample plots and we used imputation models to retrieve the forest attributes at plot and landscape-levels. The best LiDAR-derived metrics to predict HD, HM and TD were H99TH, HSD, SKE and HMIN. The Imputation model using the selected metrics was more effective for retrieving height than tree density. The model coefficients of determination (adj.R2) and a root mean squared difference (RMSD) for HD, HM and TD were 0.90, 0.94, 0.38m and 6.99, 5.70, 12.92%, respectively. Our results show that LiDAR and k-NN imputation can be used to predict stand heights with high accuracy in Pinus taeda. However, furthers studies need to be realized to improve the accuracy prediction of TD and to evaluate and compare the cost of acquisition and processing of LiDAR data against the conventional inventory procedures.