Improved Urban Scene Classification Using Full-Waveform Lidar

Abstract

Full-waveform lidar data provides supplementary radiometric as well as more accurate geometric target information, when compared to discrete return systems. In this research, a wide range of classes in an urban scene; including trees, medium vegetation, low vegetation (grass), water bodies, pitched roofs, flat roofs, asphalt, vehicles, power lines, walls (fences) and concrete are considered. In order to tackle the challenge of distinguishing geometrically similar classes and enhancing the separability of other targets, a new set of features based on deconvolved waveforms is introduced. The positive effect of the proposed feature dataset on classification accuracy in individual classes is shown using two ensemble classifiers (random forests and RUSBoost). Performance of the classifiers is improved by integration with sampling techniques, especially for the under-represented classes. The final output of the proposed method is a highly detailed land cover map of the urban scene, which affords good separability between critical classes. Waveform Processing In order to retrieve the target response from waveforms, a robust deconvolution method based upon sparsity-based regularization (

Imbalanced Learning in Land Cover Classification: Improving Minority Classes’ Prediction Accuracy Using the Geometric SMOTE Algorithm

“…Some of the existing remote sensing studies implement the random undersampling (RUS) method [19], which randomly reduces the number of the majority class training samples. However, this method has the disadvantage of information loss, as it discards samples from the majority class [5].…”

Terrestrial Lidar Data Classification Based on Raw Waveform Samples Versus Online Waveform Attributes

“…Nevertheless, some research studies have developed FWA techniques to derive more accurate basic features as well as some advanced features from the backscattered signal [12]. Studies have shown that employing basic and advanced waveform features along with careful radiometric calibration of the data improves multiclass classification in urban areas [12,13,29]. Furthermore, adding features such as the total number of echoes within each waveform and the position of the echo in the waveform together with some geometric and/or spectral features derived from the lidar system or integrated sensors such as multispectral or hyperspectral cameras can significantly increase the accuracy of the multi-class classification of lidar data over complex environments [5,22,30,31].…”

A Two-Step Classification Approach to Distinguishing Similar Objects in Mobile Lidar Point Clouds

“…Khoshelham et al (2013) and Weinmann et al (2015) used feature selection methods to reduce the required number of training samples. Azadbakht et al (2016) investigated different sampling strategies to overcome an imbalance in the distribution of training samples. In this paper, we propose a two-step classification method for distinguishing similar objects with insufficient and unbalanced training data.…”