Chu, H; Chen, W; Deng, L; (2024) Fine-Grained Segmentation of High-Resolution Bridge Crack Images Using Rendering Technology. In: Proceedings of the 2024 European Conference on Computing in Construction. (pp. pp. 167-174). European Council on Computing in Construction: Chania, Crete, Greece.

Preview

Text EC32024_246.pdf - Published Version Download (1MB) | Preview

Abstract

Drawing on insights from computer graphics, this study introduces the Crack Boundary Point Rendering Network (CBPRN), an innovative high-resolution (HR) image segmentation framework designed to improve UAVbased bridge crack inspections. We developed an edgeguided branch and an uneven sampling strategy, enhancing detail preservation on crack boundary areas effectively. Through comprehensive ablation experiments, the efficacy of the CBPRN was validated, demonstrating its superior performance with remarkable outcomes: a processing speed of 13.45 FPS and mIoU, mBA, and Dice scores of 87.23%, 93.56%, and 89.59%, respectively, for images beyond 2K resolution. The CBPRN establishes a new standard in HR crack image segmentation.

Download activity - last month

Export as CSVXMLJSON

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as JSONCSVXML

Archive Staff Only

View Item