|Table of Contents|

Improved BP algorithm and its application in detection of pavement crack(PDF)

长安大学学报（自然科学版）[ISSN:1006-6977/CN:61-1281/TN]

Issue:
2010年01期
Page:
46-53
Research Field:
Publishing date:
2010-02-20

Info

Title:
Improved BP algorithm and its application in detection of pavement crack
Author(s):
FAN Hai-wei1;  ZHANG Guo-yi1; 2;  DING Ai-ling1;  XIE Chang-rong1; 3;  XU Ting4
1. School of Information Engineering, Chang'an University, Xi'an 710064, Shaanxi, China; 2. School of Telecommunication Engineering, Beijing University of Posts and Telecommunications, Beijing 100876, China; 3. Yangzhou City Department of Road Management, Yangzhou 225002, Jiangsu, China; 4. Key Laboratory for Traffic Engineering of Beijing City, Beijing University of Technology, Beijing 100124, China
Keywords:
road engineering;  BP neural network;  improved BP algorithm;  pavement crack;  automatic detection
PACS:
U416.06; TP183
DOI:
-
Abstract:
Classical BP algorithm has the shortcomings such as bad anti-jamming ability, low learning rate and easily plunging into local minimum. In order to overcome these drawbacks, this paper put forward an improved BP algorithm with varying slope of activation function, presented a new kind of transfer function, designed a complex error function and adopted a new method, which can dynamically adjust different learning rate with separating layers. As a real application, the results of pavement crack detection by the improved BP algorithm are reported. The results show that: the proposed algorithm can efficiently improve detection speed by 30%; and reduces the total mean squares error 0.812 5; the new algorithm can be well applied in automatic pavement crack distress's detection. 2 tabs, 4 figs, 11 refs.

References:

[1] 肖旺新,张 雪,黄 卫.基于小波神经网络的路面破损识别[J].上海公路,2004(2):22-25. XIAO Wang-xin,ZHANG Xue,HUANG Wei.Preliminary study of pavement surface distress automation recognition based on wavelet neural network[J].Shanghai Highway,2004(2):22-25.
[2]杨志晓,冯冬青,陈铁军,等.人工神经网络技术在无损检测中的应用[J].无损检测,2002,24(6):244-249,252. YANG Zhi-xiao,FENG Dong-qing,CHEN Tie-jun,et al.Application of artificial neural networks to nondestructive testing[J].Nondestructive Testing,2002,24(6):244-249,252.
[3]王 越,曹长修.BP网络局部极小产生的原因分析及避免方法[J].计算机工程,2002,28(6):35-36,82. WANG Yue,CAO Chang-xiu.Analysis of local minimization for BP algorithm and its avoidance methods[J].Computer Engineering,2002,28(6):35-36,82.
[4]Bai Y P,Zhang H X,Hao Y L.The performance of the back propagation algorithm with varying slope of the activation function[J].Chaos,Solitons & Fractals,2007,33(7):1-9.
[5]Jiang M H,Georges G,Deng B X,et al.A fast learning algorithm for time-delay neural networks[J].Information Sciences,2002,24(3):27-39.
[6]Kamarthi S V,Pittner S.Accelerating neural network training using weight extrapolations[J].Elsevier Science on Neural Networks,2000,12(6):1285-1299.
[7]Martin F,Moller S.A scaled conjugate gradient algorithm for fast supervised learning[J].Neural Networks,2000,6(3):525-633.
[8]Yan W P,Diao C F,Tang Z N,et al.The study of gas sensor array signal processing with improved BP algorithm[J].Sensors and Actuators B,2000,66(7):283-285.
[9]丁爱玲,焦李成.基于支撑矢量机的路面破损识别[J].长安大学学报:自然科学版,2007,27(2):34-37. DING Ai-ling,JIAO Li-cheng.Automation of recogniting pavement surface distress based on support vector machine[J].Journal of Chang'an University:Natural Science Edition,2007,27(2):34-37.
[10]Cheng H D,Rong J.Novel approach to pavement cracking detection based on fuzzy set theory[J].Journal of Computing in Civil Engineering,1999,13(4):270-280.
[11]张 娟,沙爱民,孙朝云,等.基于相位编组法的路面裂缝自动识别[J].中国公路学报,2008,21(2):39-42. ZHANG Juan,SHA Ai-min,SUN Zhao-yun,et al.Pavement crack automatic recognition based on phase-grouping method[J].China Journal of Highway and Transport,2008,21(2):39-42.

Memo

Memo:
-
Last Update: 2010-02-20