Evolutionary feature selection to estimate forest stand variables using LiDAR

García-Gutiérrez, Jorge; González-Ferreiro, Eduardo; Santos, José Cristóbal Riquelme; Miranda, David; Diéguez-Aranda, Ulises; Navarro-Cerrillo, Rafael M.

doi:10.1016/j.jag.2013.06.005

Cited by 32 publications

(30 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Genetic approaches have made, for instance, a good variable selection in studies related to LiDAR for estimation of forest stand variables (García-Gutiérrez et al, 2013). In our case, for multiple regression methods two criteria, the Akaike Information Criterion (AIC) (Akaike, 1973) and the Bayesian Information Criterion (BIC) (Schwarz, 1978) were initially tested.…”

Section: Feature Selectionmentioning

confidence: 99%

A comparative study of regression methods to predict forest structure and canopy fuel variables from LiDAR full-waveform data

2016

View full text Add to dashboard Cite

Regression methods are widely employed in forestry to predict and map structure and canopy fuel variables. We present a study where several regression models (linear, non-linear, regression trees and ensemble) were assessed. Independent variables were calculated using metrics extracted from full-waveform LiDAR data, while the reference data used to generate the dependent variables for the prediction models were obtained from fieldwork in 78 plots of 16 m radius. Transformations of dependent and independent variables with feature selection were carried out to assess their influence in the prediction of response variables. In order to evaluate significant differences and rank regression models we used the non-parametric tests Wilcoxon and Friedman, and post-hoc analysis or post-hoc pairwise multiple comparison tests, such as Nemenyi, for Friedman test. Regressions using transformation of the dependent variable, like square-root or logarithmic, or the independent variable, increased R 2 up to 6% with respect to linear regression using unprocessed response variables. CART (Classification and Regression Tree) method provided poor results, but it may be interesting for categorisation purposes. Square-root transformation of the dependent variable is the method having the best overall results, except for stand volume. However, not always has a significant improvement with respect to other regression methods.Key words: regression models, Random Forest, CART, M5, Wilcoxon, Friedman, forest structure, canopy fuel, LiDAR full-waveform. Estudio comparativo de métodos de regresión para la predicción de variables de estructura y combustibilidad a partir de datos LiDAR full-waveformResumen: Los métodos de regresión se utilizan ampliamente en el ámbito forestal para la predicción y el cartografiado de las variables de estructura y combustibilidad. En este artículo se evalúan diferentes modelos de regresión (lineal, no lineal, árboles de regresión y ensemble). Como variables independientes se utilizaron métricas extraídas de datos LiDAR full-waveform, mientras que los valores de las variables dependientes se generaron a partir de modelos basados en datos de campo obtenidos para 78 parcelas de 16 m de radio. Se llevaron a cabo transformaciones de las variables dependientes e independientes con selección de atributos para evaluar su influencia en la predicción de la variable respuesta. Con el fin de verificar diferencias significativas y ordenar los modelos de regresión se emplearon los tests no paramétricos de Wilcoxon y Friedman, y el análisis post-hoc o los tests de comparación post-hoc por pares, como el de Nemenyi, para el test de Friedman. Las regresiones basadas en la transformación de la variable dependiente,

show abstract

Section: Feature Selectionmentioning

confidence: 99%

A comparative study of regression methods to predict forest structure and canopy fuel variables from LiDAR full-waveform data

2016

View full text Add to dashboard Cite

show abstract

“…Once the stepwise model results were available, we proceeded with the GA selection techniques (Renner, Ekárt 2003;García-Gutiérrez et al 2014). The GA selection used in this paper was implemented using the Watchmaker framework (García-Gutiérrez et al 2014). The goodness of fit of each individual model, i.e.…”

Section: P Sylvestrismentioning

confidence: 99%

“…The main concerns involve the dependence of results on the order of parameter entry (or deletion) and the absence of the control of the inflation of Type I errors in the sequence of statistical tests (Whittingham et al 2006). Recently, some authors have proposed to overcome the classic limitations of stepwise MLR on LiDAR by using evolutionary techniques (Latifi et al 2010;García-Gutiérrez et al 2014). Specifically, genetic algorithm (GA) selection appears to be a suitable tool to overcome the above-mentioned problems related to variable selection.…”

mentioning

confidence: 99%

Impact of plot size and model selection on forest biomass estimation using airborne LiDAR: A case study of pine plantations in southern Spain

et al. 2017

View full text Add to dashboard Cite

Navarro-Cerrillo R.M., González-Ferreiro E., García-Gutiérrez J., Ceacero Ruiz C.J., Hernández-Clemente R. (2017): Impact of plot size and model selection on forest biomass estimation using airborne LiDAR: A case study of pine plantations in southern Spain. J. For. Sci., We explored the usefulness of LiDAR for modelling and mapping the stand biomass of two conifer species in southern Spain. We used three different plot sizes and two statistical approaches (i.e. stepwise selection and genetic algorithm selection) in combination with multiple linear regression models to estimate biomass. 43 predictor variables derived from discrete-return LiDAR data (4 pulses per m 2 ) were used for estimating the forest biomass of Pinus sylvestris Linnaeus and Pinus nigra Arnold forests. Twelve circular plots -six for each species -and three different fixed-radius designs (i.e. 7, 15, and 30 m) were established within the range of the airborne LiDAR. The Bayesian information criterion and R 2 were used to select the best models.As expected, the models that included the largest plots (30 m) yielded the highest R 2 value (0.91) for Pinus sp. using genetic algorithm models. Considering P. sylvestris and P. nigra models separately, the genetic algorithm approach also yielded the highest R 2 values for the 30-m plots (P. nigra: R 2 = 0.99, P. sylvestris: R 2 = 0.97). The results we obtained with two species and different plot sizes revealed that increasing the size of plots from 15 to 30 m had a low effect on modelling attempts.

show abstract

“…For regression, we propose a hybrid approach based on joint use of the common-in-literature multiple linear regression (proved valid in a small-size context such as plot-level datasets) and a genetic algorithm for the selection of the best LiDAR statistics [4] in the context of the estimation of forest variables. Genetic selection provides an affordable solution in terms of computational cost in contrast with an exhaustive search where the increasing number of statistics may be an important problem and the classical stepwise regression.…”

Section: Lidar Regressionmentioning

confidence: 99%

Improving models for environmental applications of LiDAR: Novel approaches based on soft computing

García-Gutiérrez

2014

AIC

Self Cite

View full text Add to dashboard Cite

Abstract. This work proposes novel methodologies to improve the use of Light Detection And Ranging (LiDAR) for environmental purposes, especially for thematic mapping (LiDAR only or fused with other remote sensors) and the estimation of forest variables. The methodologies make use of well-known techniques from soft computing (machine learning and evolutionary computation) and their adaptation to develop LiDAR-derived products.

show abstract

Evolutionary feature selection to estimate forest stand variables using LiDAR

Cited by 32 publications

References 39 publications

A comparative study of regression methods to predict forest structure and canopy fuel variables from LiDAR full-waveform data

A comparative study of regression methods to predict forest structure and canopy fuel variables from LiDAR full-waveform data

Impact of plot size and model selection on forest biomass estimation using airborne LiDAR: A case study of pine plantations in southern Spain

Improving models for environmental applications of LiDAR: Novel approaches based on soft computing

Contact Info

Product

Resources

About