基于深度学习的排水管道缺陷自动检测与分类

Automatic detection and classification of sewer defects via deep learning

传统排水管道检测方法采用手动设计的特征,不能全面地显示这些缺陷,检测效率低且容易出错。为此,采用深度学习的方法研究排水道图像缺陷的特征表示,进而提出了一种基于两级分层深度卷积神经网络的排水道缺陷自动识别与分类系统。使用一个超过40000张排水道管道图像的训练集对网络模型进行训练。结果表明,该方法对6类管道缺陷的分类准确率超过90%,与传统方法相比大幅提高了分类精度。同时该系统现已成功应用于实际生产,验证了其在实际应用中的鲁棒性和可行性。

Video and image sources are frequently applied in the area of defect inspection in industrial community.For the recognition and classification of sewer defects,a significant number of videos and images of sewers are collected.These data are then checked by human and some traditional methods to recognize and classify the sewer defects,which is inefficient and error-prone.Previously developed features are unable to comprehensively represent such defects.In this paper,an automatic extraction of feature representation for sewer defects via deep learning is studied.Moreover,a complete automatic system for classifying sewer defects is proposed built on a 2-level hierarchical deep convolutional neural network,which shows high performance with respect to classification accuracy over 90%.The proposed network is trained on a novel dataset with over 40000sewer images.The system has been successfully applied in the practical production,confirming its robustness and feasibility to real-world applications.