• Complex
  • Title
  • Keyword
  • Abstract
  • Scholars
  • Journal
  • ISSN
  • Conference
搜索

Author:

Zhang, Allen (Zhang, Allen.) | Wang, Kelvin C. P. (Wang, Kelvin C. P..) | Fei, Yue (Fei, Yue.) | Liu, Yang (Liu, Yang.) | Chen, Cheng (Chen, Cheng.) | Yang, Guangwei (Yang, Guangwei.) | Li, Joshua Q. (Li, Joshua Q..) | Yang, Enhui (Yang, Enhui.) | Qiu, Shi (Qiu, Shi.)

Indexed by:

EI Scopus SCIE

Abstract:

A recurrent neural network (RNN) called CrackNet-R is proposed in the article for fully automated pixel-level crack detection on three-dimensional (3D) asphalt pavement surfaces. In the article, a new recurrent unit, gated recurrent multilayer perceptron (GRMLP), is proposed to recursively update the internal memory of CrackNet-R. Unlike the widely used long short-term memory (LSTM) and gated recurrent unit (GRU), GRMLP is intended for deeper abstractions on the inputs and hidden states by conducting multilayer nonlinear transforms at gating units. CrackNet-R implements a two-phase sequence processing: sequence generation and sequence modeling. Sequence generation is specifically developed in the study to find the best local paths that are most likely to form crack patterns. Sequence modeling predicts timely probabilities of the input sequence being a crack pattern. In terms of sequence modeling, GRMLP slightly outperforms LSTM and GRU by using only one more nonlinear layer at each gate. In addition to sequence processing, an output layer is proposed to produce pixel probabilities based on timely probabilities predicted for sequences. The proposed output layer is critical for pixel-perfect accuracy, as it accomplishes the transition from sequence-level learning to pixel-level learning. Using 3,000 diverse 3D images, the training of CrackNet-R is completed through optimizing sequence modeling, sequence generation, and the output layer serially. The experiment using 500 testing pavement images shows that CrackNet-R can achieve high Precision (88.89%), Recall (95.00%), and F-measure (91.84%) simultaneously. Compared with the original CrackNet, CrackNet-R is about four times faster and introduces tangible improvements in detection accuracy.

Keyword:

Author Community:

  • [ 1 ] [Zhang, Allen]Oklahoma State Univ, Sch Civil & Environm Engn, Stillwater, OK 74078 USA
  • [ 2 ] [Wang, Kelvin C. P.]Oklahoma State Univ, Sch Civil & Environm Engn, Stillwater, OK 74078 USA
  • [ 3 ] [Fei, Yue]Oklahoma State Univ, Sch Civil & Environm Engn, Stillwater, OK 74078 USA
  • [ 4 ] [Liu, Yang]Oklahoma State Univ, Sch Civil & Environm Engn, Stillwater, OK 74078 USA
  • [ 5 ] [Chen, Cheng]Oklahoma State Univ, Sch Civil & Environm Engn, Stillwater, OK 74078 USA
  • [ 6 ] [Yang, Guangwei]Oklahoma State Univ, Sch Civil & Environm Engn, Stillwater, OK 74078 USA
  • [ 7 ] [Li, Joshua Q.]Oklahoma State Univ, Sch Civil & Environm Engn, Stillwater, OK 74078 USA
  • [ 8 ] [Wang, Kelvin C. P.]Southwest Jiaotong Univ, Sch Civil Engn, Chengdu, Sichuan, Peoples R China
  • [ 9 ] [Yang, Enhui]Southwest Jiaotong Univ, Sch Civil Engn, Chengdu, Sichuan, Peoples R China
  • [ 10 ] [Qiu, Shi]Beijing Univ Technol, Sch Transportat Engn, Beijing, Peoples R China

Reprint Author's Address:

  • [Wang, Kelvin C. P.]Oklahoma State Univ, Sch Civil & Environm Engn, Stillwater, OK 74078 USA;;[Wang, Kelvin C. P.]Southwest Jiaotong Univ, Sch Civil Engn, Chengdu, Sichuan, Peoples R China

Show more details

Related Keywords:

Related Article:

Source :

COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING

ISSN: 1093-9687

Year: 2019

Issue: 3

Volume: 34

Page: 213-229

9 . 6 0 0

JCR@2022

ESI Discipline: ENGINEERING;

ESI HC Threshold:136

JCR Journal Grade:1

Cited Count:

WoS CC Cited Count: 226

SCOPUS Cited Count: 249

ESI Highly Cited Papers on the List: 17 Unfold All

  • 2024-11
  • 2024-11
  • 2024-9
  • 2024-9
  • 2024-7
  • 2024-5
  • 2024-3
  • 2024-1
  • 2023-11
  • 2023-9
  • 2023-7
  • 2023-5
  • 2023-3
  • 2023-1
  • 2022-11
  • 2022-9
  • 2022-7

WanFang Cited Count:

Chinese Cited Count:

30 Days PV: 1

Online/Total:732/5413888
Address:BJUT Library(100 Pingleyuan,Chaoyang District,Beijing 100124, China Post Code:100124) Contact Us:010-67392185
Copyright:BJUT Library Technical Support:Beijing Aegean Software Co., Ltd.