BoundaryNet Extraction and Completion of Road Boundaries With Deep Learning Using Mobile Laser Scanning Point Clouds and Satellite Imagery
ABSTARCT :
Robust road boundary extraction and completion play an important role in providing guidance to all road users and supporting high-definition (HD) maps. The significant challenges remain in remarkable and accurate road boundary recovery from poor road boundary conditions. This paper presents a novel deep learning framework, named BoundaryNet, to extract and complete road boundaries by using both mobile laser scanning (MLS) point clouds and high-resolution satellite imagery. First, road boundaries are extracted by conducting a curb-based extraction method. Such extracted 3D road boundary lines are used as inputs to feed into a U-shaped network for erroneous boundary denoising.Then, a convolutional neural network (CNN) model is proposed to complete the road boundaries. Next, to achieve more complete and accurate road boundaries, a conditional deep convolutional generative adversarial network (c-DCGAN) with the assistance of road centerlines extracted from satellite images is developed. Finally, according to the completed road boundaries, the inherent road geometries are calculated. The proposed methods were evaluated using satellite imagery and four MLS point cloud datasets with varying densities and road conditions in urban environments. The quality evaluation metrics of 82.88%, 82.43%, 88.86%, and 84.89% were achieved for four data sets. The experimental results indicate that the BoundaryNet model can provide a promising solution for road boundary completion and road geometry estimation.
EXISTING SYSTEM :
? Compared to existing point cloud segmentation methods based on traditional CNNs, the proposed method is less sensitive to data distribution and computational powers.
? The experimental results acquired by using different point cloud scenarios indicate that the MS-PCNN model can achieve state-of-the-art semantic segmentation performance in feature representativeness, segmentation accuracy, and technical robustness.
? Although such methods have significantly improved the accuracy of road boundary extraction by introducing spectral or texture information, they still cannot address existing gaps in road boundaries.
DISADVANTAGE :
? Although remarkable improvement has been achieved, most of the off-the-shelf methods cannot accurately and completely extract road boundaries, due to occlusions and point density variations during raw MLS data acquisition .
? Thus, in this paper, we mainly focus on the theoretical and methodological problems of road boundary extraction and completion using MLS point clouds and satellite imagery.
? Such max-pooling indices can significantly decrease the number of parameters facilitating end-to-end training. The dropout operation is implemented to reduce the over-fitting problem.
PROPOSED SYSTEM :
• To address this, a capsule-based deep learning framework is proposed for road marking extraction and classification from massive and unstructured MLS point clouds.
• The innovation of this study is to demonstrate the practical application of combing capsule networks with hierarchical feature encodings of georeferenced feature images for updating road information and supporting HD maps.
• The novel architecture of the proposed neural network is to directly consume unstructured 3D points and implement a point-wise semantic label assignment network to learn fine-grained layers of feature representations and reduce unnecessary convolutional computations.
ADVANTAGE :
? The proposed c-DCGAN model delivers better completion performance on the missing parts of smaller sizes. Still, if the grid cell size is too large, the generated road boundary images are in low-quality and coarse resolutions, resulting in incorrect gap detection and completion results.
? Although the completeness and curvatures vary in these road scenarios, the experimental results indicate that the BoundaryNet obtains good road boundary completion performance.
? To estimate the influence of high-resolution satellite imagery, we further evaluated the BoundaryNet completion performance with or without the assistance of road centerlines. More specifically, 150 epochs were run on the HCIP dataset.
|