Urban Tree Classification Based on Object-Oriented Approach and Random Forest Algorithm Using Unmanned Aerial Vehicle (UAV) Multispectral Imagery

Abstract

Timely and accurate information on the spatial distribution of urban trees is critical for sustainable urban development, management and planning. Compared with satellite-based remote sensing, Unmanned Aerial Vehicle (UAV) remote sensing has a higher spatial and temporal resolution, which provides a new method for the accurate identification of urban trees. In this study, we aim to establish an efficient and practical method for urban tree identification by combining an object-oriented approach and a random forest algorithm using UAV multispectral images. Firstly, the image was segmented by a multi-scale segmentation algorithm based on the scale determined by the Estimation of Scale Parameter 2 (ESP2) tool and visual discrimination. Secondly, spectral features, index features, texture features and geometric features were combined to form schemes S1–S8, and S9, consisting of features selected by the recursive feature elimination (RFE) method. Finally, the classification of urban trees was performed based on the nine schemes using the random forest (RF), support vector machine (SVM) and k-nearest neighbor (KNN) classifiers, respectively. The results show that the RF classifier performs better than SVM and KNN, and the RF achieves the highest accuracy in S9, with an overall accuracy (OA) of 91.89% and a Kappa coefficient (Kappa) of 0.91. This study reveals that geometric features have a negative impact on classification, and the other three types have a positive impact. The feature importance ranking map shows that spectral features are the most important type of features, followed by index features, texture features and geometric features. Most tree species have a high classification accuracy, but the accuracy of Camphor and Cinnamomum Japonicum is much lower than that of other tree species, suggesting that the features selected in this study cannot accurately distinguish these two tree species, so it is necessary to add features such as height in the future to improve the accuracy. This study illustrates that the combination of an object-oriented approach and the RF classifier based on UAV multispectral images provides an efficient and powerful method for urban tree classification.

Fine Classification of UAV Urban Nighttime Light Images Based on Object-Oriented Approach

“…Our study indicated that RF had the highest classification accuracy in the three regions ( Table 5 , Table 6 and Table 7 ), which is in line with the findings of many classification studies of daytime UAV images [ 12 , 35 , 36 , 37 ]. The RF algorithm integrates multiple decision trees with good high-dimensional data processing capability and can effectively avoid noise interference.…”

“…Of course, too much feature information can lead to redundancy. Yang K et al [ 11 ] demonstrated that differences in feature dimensionality and importance are the main factors contributing to variation in olive tree extraction accuracy; Guo Q et al [ 12 ] compared different feature combination schemes and showed that the combination by feature elimination had the highest accuracy in urban tree classification. Object-oriented approaches are often combined with machine learning in the selection of classification algorithms.…”

K-means based quality prediction of object-oriented software using LR-ACO

“…The outcomes demonstrated that the projected outcomes of the model are essentially compatible with the situation. In 2022, Guo et al [36] combined an object-oriented methodology and a random forest algorithm to provide an effective and practical method for identifying urban trees. Finally, utilising the RF, SVM, and KNN classifiers, the categorization of urban trees was carried out based on the nine methods.…”

Use of unmanned aerial vehicle-derived multi-spectral data for the early detection of multi-temporal maize leaf equivalent water thickness and fuel moisture content for the improved resilience of smallholder maize farming

“…34,35 The RF algorithm has high computational efficiency in processing non-parametric and high dimensional data while being sensitive to overfitting. [35][36][37] The success of the RF regression model has been well-documented in recent literature. 38,39 A comparative study by Lee et al 40 assessed the performance of SVM, multiple linear regressions, and RF algorithms in predicting the canopy nitrogen content of maize.…”