摘要
桥梁裂缝检测是桥梁养护作业中的一项重要工作,但目前的裂缝检测效率和质量仍有待提高,难以满足未来大量桥梁检测的任务需求。为解决传统检测存在的桥梁裂缝识别效率低、效果不佳等问题,提出一种基于深度学习YOLOv5s的改进算法,实现对桥梁裂缝的识别与分类。对数据集进行分割与数据增强处理,再采用Labelimg图像标注软件制作裂缝分割模型训练数据集,构建自注意力机制(CBAM)增强模型。实验结果表明,所提出的裂缝检测模型能够实现对桥梁裂缝高精度、高效率、智能化的检测,具有较强的研究价值和广泛的应用前景。
Bridge crack detection is an important task in bridge maintenance operations.However,the current efficiency and quality of crack detection still need to be improved,which has brought difficulties to meet the needs of a large number of future bridge detection tasks.In order to solve the problems of low efficiency and poor effectiveness in identifying bridge cracks in traditional detection,an improved algorithm based on deep learning YOLOv5s is proposed to achieve the recognition and classification of bridge cracks.The dataset is segmented and enhanced,Labelimg image annotation software is used to create a crack segmentation model to train the dataset,and construct a self attention mechanism enhancement model.The experimental results show that the proposed crack detection model can achieve high-precision,high-efficiency and intelligent detection of bridge cracks,which has strong research value and broad application prospects.
作者
彭家旭
顾亦然
PENG Jiaxu;GU Yiran(College of Automation&College of Artificial Intelligence,Nanjing University of Posts and Telecommunications,Nanjing 210230,China;Research Center of Smart Campus,Nanjing University of Posts and Telecommunications,Nanjing 210230,China)
出处
《现代电子技术》
2023年第24期135-140,共6页
Modern Electronics Technique