應用資料探勘分群分類演算法與空間資料庫技術在鋪面裂縫影像辨識之初探; Pavement Distress Image Recognition Using Clustering and Classification Algorithms and Spatial Database Techniques

NCU Institutional Repository > 工學院 > 土木工程研究所 > 博碩士論文 > Item 987654321/43331

請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/43331

題名:	應用資料探勘分群分類演算法與空間資料庫技術在鋪面裂縫影像辨識之初探;Pavement Distress Image Recognition Using Clustering and Classification Algorithms and Spatial Database Techniques
作者:	何庭武;Ting-Wu Ho
貢獻者:	土木工程研究所
關鍵詞:	路面缺陷檢測;數位影像處理;鋪面管理系統;資料探勘;Data Mining;Digital Image Processing;Pavement Management System;Pavement Distress Detection
日期:	2010-07-27
上傳時間:	2010-12-08 13:34:46 (UTC+8)
出版者:	國立中央大學
摘要:	最近幾年來汽機車的數量在世界各地快速成長，相對的公路等市區道路逐漸增加，需要定期對道路路面的狀況進行調查，路面缺陷的損壞是影響路面狀況關鍵的因素，若能在路面缺陷出現的初期就發現並追蹤其破壞狀況，從中對路面維護進行合理的決策和評估，道路的維護成本等費用將大幅的降低，但是如何在不影響車流的情況下，準確的發現並辨識路面缺陷已經成為待解決的一大難題。傳統的路面缺陷調查工作主要仰賴人工方式調查為主，其消耗的人力和時間以及對於交通的影響和調查人員的安全性問題，又無法得到客觀的評估，而造成路面缺陷檢測結果不一致等缺點，已經難以對應道路快速成長的需求，因此，為了減少人工檢測的成本與時間，將檢測過程以自動化方式來實施。本研究提出一種基於資料探勘分析與處理的自動化路面檢測方法：首先進行路面缺陷影像的資料前處理，再將影像資料轉換成分群資料的資料結構，並使用分群演算法進行路面缺陷的萃取，進而得到路面缺陷的群集資料，之後使用分類演算法辨識出缺陷的種類，並將結果儲存於裂縫資料庫。另外，本研究提出裂縫合併演算法與裂縫連結演算法針對裂縫資料庫進行相關的處理，前者將相鄰的裂縫資料合併為一個裂縫群組，後者將同一位置但不同時間的裂縫群組互相連結，經由以上兩個演算法方可記錄其裂縫在不同時間的變化，為路面檢測提供有效的依據，有利於路面維護和管理，幫助檢測人員進行合理的決策和評估。 The pavement maintenance of a road today relies mainly on manual pavement condition inspection and distress rating, and this manual method is costly, labor-intensive, time-consuming, and dangerous to construction workers and may affect traffic flow. Moreover, such the manual method is very subjective and may have a high degree of variability, being unable to provide meaningful information. Additionally, because only a small area of the road surface can be sampled by using the manual method, it may result in relatively low accuracy of pavement distress information. Hence, an automatic inspection system of pavement distress images is desired in hope of resolving the above problems. This paper presents a novel method to classify pavement distress images. The system is called pavement distress recognition system (PDRS), and it is installed on a pavement inspection vehicle with image acquiring devices. First, pavement images were processed to show only black-and-white pixels that can render true pavement cracks. Then, the pavement images were transformed into a set of clusters in order to capture the distress locations of each crack. Next, the distress types, i.e., horizontal, vertical, alligator-like, or man-hole-like, were obtained by applying a decision tree algorithm. Then, the system saves the data and images into a database and provides spatial query functions to users to retrieve crack information. Finally, the pavement distress database is generated to save the information of pavement distress, and using crack grouping algorithm (CGA) and crack linking algorithm (CLA) to create the spatial relation for distress management. The present results showed that our method can successfully recognize various types of pavement distress. Our system also provides information regarding pavement crack lifecycle, i.e., when the crack was identified, when it was fixed, etc., so that public road agencies can define maintenance plans in accordance with real pavement conditions.
顯示於類別:	[土木工程研究所] 博碩士論文

文件中的檔案:

檔案	描述	大小	格式	瀏覽次數
index.html		0Kb	HTML	683	檢視/開啟

在NCUIR中所有的資料項目都受到原著作權保護.

社群 sharing

資料載入中.....