Eric Bastos Görgens

Estimação do volume de árvores utilizando redes neurais artificiais

The artificial neural network consists of a set of units containing mathematical functions connected by weights. Such nets are capable of learning by means of synaptic weight modification, generalizing learning for other unknown archives. The neural network project comprises three stages: pre-processing, processing and post-processing of data. One of the classical problems approached by networks is function approximation. Tree volume estimate can be included in this group. Four different architectures, five pre-processings and two activation functions were used. The nets which were statistically similar to the observed data were also analyzed in relation to residue and volume and compared to the volume estimate provided by the Schumacher and Hall equation. The neural nets formed by neurons whose activation function was exponential presented estimates statistically similar to the observed data. The nets trained with the data normalized by the linear interpolation method and equalized presented better estimate performance.

Comparison of ALS based models for estimating aboveground biomass in three types of Mediterranean forest

Abstract This study aimed to develop ALS-based models for estimating stem, crown and aboveground biomass in three types of Mediterranean forest, based on low density ALS data. Two different modelling approaches were used: (i) linear models with different variable selection methods (Stepwise Selection [SS], Clustering/Exhaustive search [CE] and Genetic Algorithm [GA]), and (ii) previously Published Models (PM) applicable to diverse types of forest. Results indicated more accurate estimations of biomass components for pure Pinus pinea L. (rRMSE = 25.90-26.16%) than for the mixed (30.86-36.34%) and Quercus pyrenaica Willd. forests (32.78-34.84%). All the tested approaches were valuable, but SS and GA performed better than CE and PM in most cases.

A performance comparison of machine learning methods to estimate the fast-growing forest plantation yield based on laser scanning metrics

Machine learning models tackle high-dimensional problems.The machine learning models are not limited to a subset of predictor variables.The random forest had the best RMSE compared to neural network and support vector regression.The coefficient of determination and bias were similar to all modeling techniques. Machine learning models appear to be an attractive route towards tackling high-dimensional problems, particularly in areas where a lack of knowledge exists regarding the development of effective algorithms, and where programs must dynamically adapt to changing conditions. The objective of this study was to evaluate the performance of three machine learning tools for predicting stand volume of fast-growing forest plantations, based on statistical vegetation metrics extracted from an Airborne Laser Scanning (ALS) survey. The forests used in this study were composed of 1138ha of commercial plantations that consisted of hybrids of Eucalyptus grandis and Eucalyptus urophylla, managed for pulp production. Three machine learning tools were implemented: neural network (NN), random forest (RF) and support vector regression (SV); and their performance was compared to a regression model (RM). The RF and the RM presented an RMSE in the leave-one-out cross-validation of 31.80 and 30.56m3ha-1 respectively. The NN and SV presented a higher RMSE than the others, equal to 64.44 and 65.30m3ha-1. The coefficient of determination and bias were similar to all modeling techniques. The ranking of ALS metrics based on their relative importance for the estimation of stand volume showed some differences. Rather than being limited to a subset of predictor variables, machine learning techniques explored the complete metrics set, looking for patterns between them and the dependent variable.

Artificial Intelligence Procedures for Tree Taper Estimation within a Complex Vegetation Mosaic in Brazil

Tree stem form in native tropical forests is very irregular, posing a challenge to establishing taper equations that can accurately predict the diameter at any height along the stem and subsequently merchantable volume. Artificial intelligence approaches can be useful techniques in minimizing estimation errors within complex variations of vegetation. We evaluated the performance of Random Forest® regression tree and Artificial Neural Network procedures in modelling stem taper. Diameters and volume outside bark were compared to a traditional taper-based equation across a tropical Brazilian savanna, a seasonal semi-deciduous forest and a rainforest. Neural network models were found to be more accurate than the traditional taper equation. Random forest showed trends in the residuals from the diameter prediction and provided the least precise and accurate estimations for all forest types. This study provides insights into the superiority of a neural network, which provided advantages regarding the handling of local effects.

INFLUÊNCIA DA ARQUITETURA NA ESTIMATIVA DE VOLUME DE ÁRVORES INDIVIDUAIS POR MEIO DE REDES NEURAIS ARTIFICIAIS

Supervised neural networks are composed of parallel processing units. Each unit, called neurons, computes certain mathematical functions. The units are arranged in layers and connected by synaptic weights to balance the entries, trying to adjust them to a predetermined output pattern. The correct definition of the number of layers and the number of neurons in each layer are crucial, once the training is directly influenced by these parameters. To explore this point, data of scaling from five different regions were arranged in a spreadsheet and randomly divided into training and validation set. Data were presented for three networks with different architectures. The evaluation was performed using residual plots and t test (p <0.05). To estimate volume per tree, the neural network must be built with more than 10 neurons in the first layer, and it is recommended the use of more than one intermediate layer.

Assessing biomass based on canopy height profiles using airborne laser scanning data in eucalypt plantations

This study aimed to map the stem biomass of an even-aged eucalyptus plantation in southeastern Brazil based on canopy height profile (CHPs) statistics using wall-to-wall discrete return airborne laser scanning (ALS), and compare the results with alternative maps generated by ordinary kriging interpolation from field-derived measurements. The assessment of stem biomass with ALS data was carried out using regression analysis methods. Initially, CHPs were determined to express the distribution of laser point heights in the ALS cloud for each sample plot. The probability density function (pdf) used was the Weibull distribution, with two parameters that in a secondary task, were used as explanatory variables to model stem biomass. ALS metrics such as height percentiles, dispersion of heights, and proportion of points were also investigated. A simple linear regression model of stem biomass as a function of the Weibull scale parameter showed high correlation (adj.R2 = 0.89). The alternative model considering the 30th percentile and the Weibull shape parameter slightly improved the quality of the estimation (adj.R2 = 0.93). Stem biomass maps based on the Weibull scale parameter doubled the accuracy of the ordinary kriging approach (relative root mean square error = 6 % and 13 %, respectively).

Performance of stem denoising and stem modelling algorithms on single tree point clouds from terrestrial laser scanning

Abstract The present study assessed the performance of three different methods of stem denoising and three different methods of stem modelling on terrestrial laser scanner (TLS) point clouds containing single trees – thus validating all tested methods, which were made available as an open source software package in the R language. The methods were adapted from common TLS stem detection techniques and rely on finding one main trunk in a point cloud by denoising the data to precisely extract only stem points, followed by a circle or cylinder fitting procedure on stem segments. The combination of the Hough transformation stem denoising method and the iteratively reweighted total least squares modelling method had best overall performance – achieving 2.15 cm of RMSE and 1.09 cm of bias when estimating diameters along the stems, detecting 80% of all stem segments measured on field surveys. All algorithms performed better on point clouds of boreal species, in comparison to tropical Eucalypt. The point clouds underwent reduction of point density, which increased processing speed on the stem denoising algorithms, with little effect on diameter estimation quality.

EUCALYPTUS STAND SAMPLE PLOTS COMPARED: FIXED AREA AND FIXED NUMBER OF TREES

This study compares sampling methods based on plots of fixed area and based on a fixed number of trees. The study was conducted in a Eucalyptus forest surveyed using three plot types: rectangular with fixed area, circular with fixed area and fixed number of trees. The estimation accuracies were evaluated for the average diameter per plot and for the number of stems, the basal area and the volume per plot. The null hypothesis of equality between the sampling methods was assessed by t-test. No significant differences were found between the three sampling methods.