Abstract：To improve the efficiency of pavement disease detection and achieve the visual display of pavement disease detection results, an intelligent comprehensive pavement disease detection method of asphalt pavements was designed by combining deep learning technology and ArcMap geographic information system. The convolutional neural network based on visual geometry group (VGG) model was used for pavement image classification, and the object detection network based on single shot MultiBox detector (SSD) model was used for pavement disease identification. ArcMap mapping editing system was used to generate road health maps. The test results show that the accuracy of pavement image classification based on VGG-16 model is 94.60%, which can distinguish normal pavement and diseased pavement. The average accuracy of pavement disease recognition based on SSD model is 87.36%, which can effectively identify six types of diseases of potholes, loose, rutting, cracking, bleeding and patches. The road health map based on ArcMap system can locate the diseases and output the results.
韩豫1， 张萌1， 李宇宏1， 顾盛2. 基于深度学习和ArcMap的路面病害智能综合检测方法[J]. 江苏大学学报（自然科学版）, 2023, 44(4): 490-496.
HAN Yu1， ZHANG Meng1， LI Yuhong1， GU Sheng2. Intelligent comprehensive detection method of pavement diseases based on deep learning and ArcMap[J]. Journal of Jiangsu University(Natural Science Eidtion)
, 2023, 44(4): 490-496.
ZHOU L P, GAO M W, HE C. Study on face recognition under unconstrained conditions based on LBP and deep learning[J]. Journal of Computational Methods in Sciences and Engineering, 2021, 21(2):497-508.
WU S,WANG H T,JIANG Y, et al. Semantic segmentation algorithm of brain MRI image based on hybrid neural network[J]. Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition), 2022, 34(3):423-432.（in Chinese）
HU J,YANG H Q, DU X R, et al. Application of few-shot learning in high resolution remote sensing image classification and recognition[J]. Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition), 2022, 34(3):410-422. （in Chinese）
XU Y, WEN D S, ZHOU J P, et al. Identification method of cotton seedlings and weeds in Xinjiang based on Faster R-CNN[J]. Journal of Drainage and Irrigation Machinery Engineering, 2021, 39(6):602-607. (in Chinese)
WANG L P, GAO R Z, ZHANG J J, et al. Crack detection of concrete pavement based on convolutional neural network[J]. Computer Science, 2019,46(Sup2):584-589. (in Chinese)
［7］
赵潇. 基于深度学习的路面坑洼检测系统研究与实现[D]. 银川: 宁夏大学,2020.
［8］
SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[C]∥International Conference on Learning Representations. San Juan, Puerto Rico: Schloss Dagstuhl-Leibniz Center for Informatics, 2015:1-14.
［9］
LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot MultiBox detector[C]∥European Conference on Computer Vision. Amsterdam, the Netherlands:[s. n.], 2016:21-37.
［10］
PAN S J, YANG Q. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10):1345-1359.
FENG Y X, LI Y M. An overview of deep learning optimization methods and learning rate attenuation methods[J]. Hans Journal of Data Mining, 2018, 8(4):186-200. (in Chinese)
2023, Vol. 44 
