3D Road Boundary Extraction Based on Machine Learning Strategy Using LiDAR and Image-Derived MMS Point Clouds

Abstract

The precise extraction of road boundaries is an essential task to obtain road infrastructure data that can support various applications, such as maintenance, autonomous driving, vehicle navigation, and the generation of high-definition maps (HD map). Despite promising outcomes in prior studies, challenges persist in road extraction, particularly in discerning diverse road types. The proposed methodology integrates state-of-the-art techniques like DBSCAN and RANSAC, aiming to establish a universally applicable approach for diverse mobile mapping systems. This effort represents a pioneering step in extracting road information from image-based point cloud data. To assess the efficacy of the proposed method, we conducted experiments using a large-scale dataset acquired by two mobile mapping systems on the Yıldız Technical University campus; one system was configured as a mobile LiDAR system (MLS), while the other was equipped with cameras to operate as a photogrammetry-based mobile mapping system (MMS). Using manually measured reference road boundary data, we evaluated the completeness, correctness, and quality parameters of the road extraction performance of our proposed method based on two datasets. The completeness rates were 93.2% and 84.5%, while the correctness rates were 98.6% and 93.6%, respectively. The overall quality of the road curb extraction was 93.9% and 84.5% for the two datasets. Our proposed algorithm is capable of accurately extracting straight or curved road boundaries and curbs from complex point cloud data that includes vehicles, pedestrians, and other obstacles in urban environment. Furthermore, our experiments demonstrate that the algorithm can be applied to point cloud data acquired from different systems, such as MLS and MMS, with varying spatial resolutions and accuracy levels.

Civil Integrated Management (CIM) for Advanced Level Applications to Transportation Infrastructure: A State-of-the-Art Review

“…Three-dimensional scanning, which encompasses technologies such as LiDAR, allows for detailed measurements and high-fidelity models, particularly with terrestrial setups where the equipment is stationary. For instance, a higher-resolution sensor influences the detection capabilities, distinguishing between different types of surfaces markings, obstacles, and traffic [95][96][97]. Conversely, the spatial resolution in Remote Sensing, which involves a broader range of technologies including aerial imagery and satellite photos, can vary significantly.…”

Different KNN parameters on road surface extraction from UAV-based point cloud under several machine learning algorithms

“…Their study explicitly evaluated several training samples with different features and demonstrated a classification improvement of up to 13%. Suleymanoglu et al [30] introduced a methodology that combines techniques such as DBSCAN and RANSAC to extract road boundaries. This methodology was applied using 3D data from two distinct sources: MLS and a photogrammetry-based system.…”

“…Similar comparisons were conducted for standard deviation as well. For the accuracy measure, the standard deviations of the glm, lda, [50] 0.9300 -Ha Tran and Taweep [26] -0.9740 Hanma et al [28] -0.9070 Zeybek and Biçici [51] -0.7300 Bai et al [52] 0.6941 0.8623 Biçici and Zeybek [25] 0.9500 -Wang et al [27] -0.9200 Suleymanoglu et al [30] -0.9390 This study (study area 1) 0.9720 0.9760 This study (study area 2) 0.9590 0.9690 and linda algorithms are found to be larger than those of the rf and svm algorithms. However, this observation is the opposite for the quality measure.…”