Arboreal Identification Supported by Fuzzy Modeling for Trunk Texture Recognition

Abstract

Due to the natural variability of the arboreal bark there are texture patterns in trunk images with values belonging to more than one species. Thus, the present study analyzed the usage of fuzzy modeling as an alternative to handle the uncertainty in the trunk texture recognition, in comparison with other machine learning algorithms. A total of 2160 samples, belonging to 20 tree species from the Brazilian native deciduous forest, were used in the experimental analyzes. After transforming the images from RGB to HSV, 70 texture patterns have been extracted based on first and second order statistics. Secondly, an exploratory factor analysis was performed for dealing with redundant information and optimizing the computational effort. Then, only the first dimensions with higher cumulative variability were selected as input variables in the predictive modeling. As a result, fuzzy modeling reached a generalization ability that outperformed algorithms widely used in classification tasks, besides of obtaining an almost perfect agreement with the classifier with the best accuracy in the validation tests. Therefore, the fuzzy modeling can be considered as a competitive approach, with reliable performance in arboreal trunk texture recognition.

A Novel Data-Driven Fuzzy for Accurate Coagulant Dosage in Drinking Water Treatment

“…This algorithm develops a sequential training through which the decision trees grow in series. In this way, a tree is built to correct the errors of the previous one (boosting), which generally provides superior performance unless there is influence from noisy data (Wei et al, 2021;Bressane et al, 2018).…”

“…For the Lukasiewicz conjunction operator, 6 antecedent terms (linguistic values) were sufficient for the FIS to decrease the RMSE close to 0.44 during training. While some operators consider only the lowest membership in the disjunction step, the sum t-norm considers all membership values, which provides improved performance in the regression task(Ghodousian et al, 2018;Bressane et al, 2018).From these results, computer-aided coagulant dosage can be highly accurately determined using the FIS approach proposed in this study. As a practical implication, this alternative avoids errors associated with the WTP operator's experience; it can predict dosages accurately and in real time, saving operational resources, the acquisition and maintenance of equipment, and the consumption of raw material required by jar tests.…”

Fuzzy Machine Learning Applications in Environmental Engineering: Does the Ability to Deal with Uncertainty Really Matter?

“…Goyal et al [21] Daily Evaporation ANN, LS-SVR, FIS, ANFIS Ay and Kisi [22] COD Concentration MLR, MLP, RBF, GRNN, ANFIS, k-MLP He et al [23] River Flow ANN, ANFIS, SVM Asadi et al [24] NOx Concentration ANN, NF Tayfur et al [25] Hydraulic Conductivity SFL, MFL, LM-ANN, NF Piotrowski et al [26] Water Temperature MLP, ANFIS, WNN, KNN Olyaie et al [27] Suspended Sediment Load ANNs, ANFIS, WNN, SRC Estalaki et al [28] Water Quality ER, FSC, SWMM, MUSIC Aghbashlo et al [29] Photo-Biohydrogen Production RBF, FCR Nadiri et al [30] Strength of Geopolymers SFL, MFL, LFL Bagheri et al [31] Landfill Leachate Penetration FIS, ANN Bressane et al [32] Arboreal Recognition FIS, C5, CCNN, KNN, PNN, MLP, RF, DT, SGB, SVM Nabavi-Pelesaraei et al [33] Energy Output ANN, ANFIS Dou and Yang [34] Daily Evapotranspiration ELM, ANFIS, ANN, SVM Choubin et al [35] Suspended Sediment Load CART, ANFIS, MLP, SVM Nadiri et al [36] Effluent Water Parameters FIS, SCFL Raei et al [37] Urban Stormwater MLP, NSGA-II, Fuzzy α-cut, DSS Adnan et al [38] Daily Streamflow ANFIS-PSO, MARS, M5, OP-ELM Kaab et al [39] Environmental Impacts ANN, ANFIS Roy et al [40] Reference Evapotranspiration FA-ANFIS, Ensembles Ly et al [41] Water Quality Modeling LR, DL-ANN, ANFIS Manzar et al [42] Water Quality Index GRNN, Elm-NN, FFNN, SVM, LR, NF Kılıç and Topuz [43] PTE in Volcanic Ash Soils ANN, FLRA Recognizing the inherent uncertainties in climatic conditions, Goyal et al [21] aimed to address the challenges associated with the accurate modeling of daily evaporation predictions in subtropical climates. The methods under comparison included ANN, least squares support vector regression (LS-SVR), FIS, and adaptive neuro-fuzzy inference systems (ANFISs).…”

“…Bressane et al [32] introduced a fuzzy model for the identification of tree species, aiming to enhance accuracy in arboreal trunk texture recognition. The study compared the performance of the fuzzy-based system with several established classification algorithms, including the boosted rule-based model (C5), cascade-correlation neural network (CCNN), KNN, probabilistic neural network (PNN), MLP, random forest (RF), decision tree (DT), stochastic gradient boosting (SGB), and SVM.…”

Controle Automático de Volume em Tempo Real Utilizando Inferência Fuzzy em um Sistema Embarcado

“…A lógica Fuzzy, proposta inicialmente em 1965 [40], tem sido aplicada em diversasáreas [3,4,5,11]. O conceito fundamental da lógica Fuzzyé flexibilizar a pertinência dos elementos através do conceito de graus de pertinência [16,34], permitindo assim modelar a incerteza e a subjetividade de sistemas complexos.…”