Chu, Honghu; Chen, Weiwei; Deng, Lu; (2025) Probability Map Guided Point Rendering Technique for Refined Segmentation of High-Resolution Crack Images. In: Cheng, Jack CP and Yabuki, Nobuyoshi and Yu, Yantao and ICCBEI 2025 Organizing Committee, (eds.) Proceedings of the Sixth International Conference on Civil and Building Engineering Informatics (ICCBEI 2025). (pp. pp. 666-675). Hong Kong University of Science and Technology: Hong Kong, China.

Text ICCBEI_2025_paper_47- Probability Map Guided Point Rendering Technique for Refined Segmentation of High-Resolution Crack Images- Accepted version.pdf - Accepted Version Access restricted to UCL open access staff until 6 May 2025. Download (817kB)

Abstract

High-resolution (HR) imaging devices are now widely used for capturing crack images from civil structures, necessitating the development of algorithms for HR image segmentation. However, the traditional refined segmentation of HR images requires substantial GPU resources, which leads to the adoption of the costeffective point rendering technique for inference. Considering that traditional rendering techniques require the use of coarse masks to guide the rendering points for processing prediction, these coarse masks typically fail to effectively focus the rendering points on the boundary regions of the slender cracks, resulting in ambiguous predictions at crack boundaries. In contrast, we introduce a novel rendering point sampling paradigm that enables the network to focus rendering points on crack boundary regions, guided by the probability maps during the inference phase. This approach significantly improves the segmentation accuracy of crack boundary regions from HR images without increasing computational resource dependence. Experiments on an open-source HR crack image dataset consistently show our method's superiority over state-of-the-art approaches, with final results of 84.24%, 93.78%, and 91.45% on IoU, mBA, and Dice, respectively.

Download activity - last month

Export as XMLCSVJSON

Download activity - last 12 months

Export as XMLJSONCSV

Downloads by country - last 12 months

Export as CSVJSONXML

Archive Staff Only

View Item