Now 5709 visitors
Today:0 Yesterday:0
Total: 148879
212S 50P 4R
2026-10-20, Week 43 |
| Member Login |
| TACT Journal Homepage |
| Call for Paper |
| Committee |
| Find My Paper |
| Author Homepage |
| Statistics & History |
| Welcome Message |
| Paper Procedure |
| FAQ |
| Paper Archives |
| Outstanding Papers |
| Author Homepage |
| - Paper Procedure |
| - Journal Procedure |
| - Presentation Tips |
|
| Program & Proceedings |
| Program with Papers |
| Plenary Session |
| Tutorial Session |
|
| Presentation Platform |
| Hotel & Travel Info |
| About Korea |
| Accommodation |
| Transportation |
| VISA |
| Other Infomation |
|
| Photo Gallery |
| Scheduler Login |
| Seminar |
| Archives Login |
|
| Sponsors |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Save Q&A |
 |
* Edit or answer any Q&A by selecting Q&A number Hyper Link below + Write button (Save)
|
ICACT20260211 Question.1
Questioner: quandh13@fe.edu.vn 2026-02-11 오후 10:25:09 |
ICACT20260211 Answer.1
Answer by Auhor taherul203158@gmail.com 2026-02-11 오후 10:25:09
Chrome  Click!! |
| The authors present a study on road surface macro-texture classification. While some expected outcomes are described, at least some of the current results should be clarified, particularly regarding the datasets that were collected and used in the experiments. In addition, could the authors clarify whether multiple distinct types of road cracks are considered in this work? |
Thank you for your question. The present study primarily concentrates on the development of a machine vision–based framework for road surface macro-texture measurement and classification, rather than on the categorization of specific pavement crack types. As detailed in the methodology section, the dataset comprises high-resolution pavement surface images captured using vehicle-mounted camera systems. In certain experimental configurations, stereo vision or laser triangulation technologies are additionally employed to reconstruct three-dimensional surface profiles and enhance texture depth representation. All images are acquired under real-world field conditions and subsequently processed through a series of preprocessing procedures, including noise reduction, contrast enhancement, and edge detection techniques such as the Sobel and Canny operators. These steps are performed to improve feature visibility and ensure the robustness of subsequent texture feature extraction. The processed data are then utilized for macro-texture characterization and classification through a convolutional neural network (CNN)-based framework. Importantly, the classification objective of this study is to distinguish between different levels of macro-texture or surface roughness, rather than to identify or categorize specific pavement distress types. Although crack detection is discussed in the literature review as one of the earliest and most widely explored applications of machine vision in pavement engineering, it is included solely to provide contextual background and demonstrate the evolution of related technologies. Accordingly, this research does not treat different crack types (e.g., longitudinal, transverse, or alligator cracking) as separate classification categories. While cracks may incidentally appear in some images and influence texture measurements, particularly in cases such as MPD outliers, they are regarded as surface irregularities affecting macro-texture characteristics, rather than as independent analytical targets. Crack-type differentiation, therefore, falls outside the defined scope and objectives of the proposed framework. |
| |
