A Convolutional-Transformer Network for Crack Segmentation with Boundary Awareness

• Published in: 2023 IEEE International Conference on Image Processing (ICIP) pp. 86 - 90
• Main Authors: Tao, Huaqi, Liu, Bingxi, Cui, Jinqiang, Zhang, Hong
• Format: Conference Proceeding
• Language: English
• Published: IEEE 08.10.2023
• Subjects: Architecture; Buildings; Convolutional codes; Convolutional Neural Network; Crack Segmentation; Deep Learning Application; Image segmentation; Safety; Task analysis; Transformers; Vision Transformer

Cracks play a crucial role in assessing the safety and durability of manufactured buildings. However, the long and sharp topological features and complex background of cracks make the task of crack segmentation extremely challenging. In this paper, we propose a novel convolutional-transformer networ...