An application of R Software model based on Deep Learning Algorithms to future usage of other forest practitioner for predicting individual tree height

In this study, the artificial intelligence models based on Deep learning Algorithms were developed to model the relationships between the individual tree total heights (ITH) and diameter at breast heights (DBH) with the stand variables. The H20 package, which have been coded in R software language, with an h2o.deeplearning function, which was coded in Java, was used to train these DLA models and obtain the ITH predictions. To determine best predictive input variables, various input variable alternatives were evaluated based on the statistical fitting criteria. From these fitting statistics for the training data set, the DLA model which includes the input variables with the DBH, dominant diameter (cm), dominant height (m), number of trees in hectare and basal area (m2/ha) resulted in the best predictive statistics with a RMSE value of 0.7173, RMSE% value of 4.5986, the AIC value of -291.3037, BIC value of 1158.4564, FI of 0.9785 values, AAE value of 0.4311, Bias value of 0.0438 and Bias% value of 0.2805. Similar to the fitting statistics in training data, the DLA model which includes the input variables with the DBH, dominant diameter (cm), dominant height (m), number of trees in hectare and basal area (m2/ha) gave the best predictive statistics with a RMSE value of 1.8217, RMSE% value of 10.2151, the AIC value of 99.9615, BIC value of 331.3772, FI of 0.8334 values, AAE value of 1.2051, Bias value of -0.0985 and Bias% value of -0.5521. To train these DLA models, R software platform, which is free and open for all, was used to share with various stakeholders and other users in forest management. Thus, besides the modeling studies including the comparison of various network models with classical regression models, the opportunity to other forest practitioner to use artificial intelligence model developed in this study can be achieved by downloading this best predictive DLA model from the supplementary file section of this study.

Anahtar Kelimeler:

Individual Tree Height, Deep Learning Algorithms, H20 package, h2o.deeplearning function

PDF

___

Carranza-Rojas, J., Goeau, H., Bonnet, P., Mata-Montero, E., Joly, A. 2017. Going deeper in the automated identification of Herbarium specimens. BMC Evolutionary Biology 17, 181.Clutter, J. L., Fortson, J. C., Pienaar, L. V., Brister, G. H., Bailey, R. L. 1983. Timber management: a quantitative approach. John Wiley & Sons, Inc.Ercanlı, İ. 2015. Nonlinear mixed effect models for predicting relationships between total height and diameter of oriental beech trees in Kestel, Turkey. – Revista Chapingo Serie Ciencias Forestales y del Ambiente 21(1): 185-202. Ferentinos, K.P., 2018. Deep learning models for plant disease detection and diagnosis. Computers and Electronics in Agriculture 145, 311-318.Gadow, K. V. and Hui, G. Y. 1999. Modelling Forest Development, Kluwer Academic Publishers, Dordrect, 213 s.Huang, S., Titus, S. J., Wiens, D. P. 1992. Comparison of nonlinear heightdiameter functions for major Alberta tree species. Canadian Journal of Forest Research, 22, 1297–1304.H2O.ai Team. H2O GitHub Repository, 2018. URL: https://github.com/h2oaiLee, S. H., Chan, C. S., Wilkin, P., Remagnino, P. 2015. Deep-plant: Plant identification with convolutional neural networks. In, 2015 IEEE International Conference on Image Processing (ICIP). IEEE, pp. 452-456.Martin, F. C., Flewelling, J. W. 1998. Evaluation of tree height prediction models for stand inventory. Western Journal of Applied Forestry, 13, 109–119.Mohanty, S. P., Hughes, D. P., Salathé, M. 2016. Using deep learning for image-based plant disease detection. Frontiers in plant science 7, 1419.Nanos, N., Calama, R., Montero, G., Gil, L. 2004. Geostatistical prediction of height-diameter models. Forest Ecology and Management, 195, 221-235.Poudel, K. P., Cao, Q. V. 2013. Evaluation of methods to predict Weibull parameters for characterizing diameter distributions. Forest Science 59: 243-252.R Core Team (2018) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna., Austria.: https://www.R-project.orgSAS Institute Inc. 2004. SAS/IML 9.3 User’s Guide. Cary, USA. – SAS Institute Inc., Cary, NC, USASladojevic, S., Arsenovic, M., Anderla, A., Culibrk, D., Stefanovic, D. 2016. Deep neural networks based recognition of plant diseases by leaf image classification. Computational Intelligence Neuroscience 2016.Schnute, J., 1981. A versatile growth model with statistically stable parameters. Canadian Journal of Fisheries Aquatic Sciences 38, 1128-1140.Sun, Y., Liu, Y., Wang, G., Zhang, H. 2017. Deep learning for plant identification in natural environment. Computational Intelligence Neuroscience 2017.van Laar, A., Akça, A. 2007. Forest mensuration: in Managing Forest Ecosystems, Dordrecht, The Netherlands: Springer. 383 p.Ubbens, J., Cieslak, M., Prusinkiewicz, P., Stavness, I., 2018. The use of plant models in deep learning: an application to leaf counting in rosette plants. Plant methods 14, 6.Zeiler M.D., 2012. ADADELTA: an adaptive learning rate method. ArXiv-Machine Learning.