当前位置: X-MOL 学术Autom. Constr. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Multimodal deep learning-based automatic generation of repair proposals for steel bridge shallow damage
Automation in Construction ( IF 9.6 ) Pub Date : 2025-01-13 , DOI: 10.1016/j.autcon.2025.105961
Honghong Song, Xiaofeng Zhu, Haijiang Li, Gang Yang

As bridges age, manual repair decision-making methods struggle to meet growing maintenance demands. This paper develops AI systems that can imitate experts' decision processes by mining implicit relationships between bridge damage images and corresponding repair proposals. A multimodal deep learning-based end-to-end decision-making method is proposed to extract and map features of bridge damage images and repair proposal texts, automating damage repair proposal generation. The model is trained and validated using a dataset from historical inspection reports. The model's image feature extraction is evaluated using Class Activation Mapping (CAM), while text generation achieved BLEU-1 to BLEU-4 scores of 0.76, 0.743, 0.712, and 0.705, respectively, with 82 % accuracy in human evaluation. The results indicate the model's effectiveness in handling complex image features and generating long text, addressing challenges in automated bridge repair decision-making.

中文翻译:


基于多模态深度学习的钢桥浅层损伤自动生成修复方案



随着桥梁的老化,手动维修决策方法难以满足不断增长的维护需求。本文开发了人工智能系统,该系统可以通过挖掘桥梁损伤图像和相应修复方案之间的隐含关系来模仿专家的决策过程。提出了一种基于多模态深度学习的端到端决策方法,用于提取和映射桥梁损伤图像和修复方案文本的特征,从而自动生成损伤修复方案。该模型使用历史检查报告中的数据集进行训练和验证。使用类激活映射 (CAM) 评估模型的图像特征提取,而文本生成则实现了 BLEU-1 到 BLEU-4 的分数分别为 0.76、0.743、0.712 和 0.705,在人工评估中准确率为 82%。结果表明,该模型在处理复杂图像特征和生成长文本方面的有效性,解决了自动化桥梁维修决策中的挑战。
更新日期:2025-01-13
down
wechat
bug