A Two-Step Feature Extraction Algorithm: Application to Deep Learning for Point Cloud Classification

Abstract

Abstract. Most deep learning (DL) methods that are not end-to-end use several multi-scale and multi-type hand-crafted features that make the network challenging, more computationally intensive and vulnerable to overfitting. Furthermore, reliance on empirically-based feature dimensionality reduction may lead to misclassification. In contrast, efficient feature management can reduce storage and computational complexities, builds better classifiers, and improves overall performance. Principal Component Analysis (PCA) is a well-known dimension reduction technique that has been used for feature extraction. This paper presents a two-step PCA based feature extraction algorithm that employs a variant of feature-based PointNet (Qi et al., 2017a) for point cloud classification. This paper extends the PointNet framework for use on large-scale aerial LiDAR data, and contributes by (i) developing a new feature extraction algorithm, (ii) exploring the impact of dimensionality reduction in feature extraction, and (iii) introducing a non-end-to-end PointNet variant for per point classification in point clouds. This is demonstrated on aerial laser scanning (ALS) point clouds. The algorithm successfully reduces the dimension of the feature space without sacrificing performance, as benchmarked against the original PointNet algorithm. When tested on the well-known Vaihingen data set, the proposed algorithm achieves an Overall Accuracy (OA) of 74.64% by using 9 input vectors and 14 shape features, whereas with the same 9 input vectors and only 5PCs (principal components built by the 14 shape features) it actually achieves a higher OA of 75.36% which demonstrates the effect of efficient dimensionality reduction.

Resampling Methods for a Reliable Validation Set in Deep Learning Based Point Cloud Classification

“…To understand the developed model, and the classification results for the ALS data, we use five common evaluation metrics: F1-score (F1), mean F1 (mF1), Intersection over Union (IoU), mean IoU (mIoU), and the Overall Accuracy (OA). The reader is referred to Nurunnabi et al (2022) for detail about the metrics.…”

“…We use the spatial coordinates as well as the LiDAR features (I and RN), and heights as the raw inputs. The heights are the differences between the z values of a point of interest and the lowest point in the local neighbourhood of the interest point (Nurunnabi et al, 2022). Hence, individual points are characterized by their coordinates (x, y, z), I, RN, and heights.…”

Robust Approach for Urban Road Surface Extraction Using Mobile Laser Scanning 3d Point Clouds

“…Unlike images, laser scanning point clouds offer detailed 3D geometry including distance information of objects. But, analysis of point clouds is not very easy as point clouds are usually unstructured, have inhomogeneous point density and irregular data format, and are typically capture sharp features (e.g., edges and corners) and arbitrary surface shapes (Nurunnabi et al, 2022). Additionally, point clouds are not free from the presence of noise, outliers and occlusions.…”