Automatic concrete sidewalk deficiency detection and mapping with deep learning

Vertical displacement is a common concrete slab sidewalk deficiency, which may cause trip hazards and reduce wheelchair accessibility. This paper presents an automatic approach for hazard detection mapping based on deep learning. A low-cost mobile LiDAR scanner was used to obtain full-width as-is conditions of sidewalks, after method developed convert the scanned 3D point clouds into 2D RGB orthoimages elevation images. Then, learning-based model pixelwise segmentation joints. Algorithms were extract different types joints straight curved sidewalks from segmented evaluated by measuring differences adjacent edges parallel boundaries joints, potential identified. In end, detected normal geo-visualized with specific information Web GIS. Experiments demonstrated proposed performed well segmenting images, highest IoU (Intersection over Union) 0.88, achieved similar results compared manual assessment detecting but higher efficiency. The cost- time-effective, expected enhance improve safety general public.