| 摘要: | 鋪面狀況指標為鋪面重要的服務指數，不僅影響乘車的舒適度，亦提供於鋪面建設與維護評估之時機，而如何有效管理道路狀況，需定期檢驗及蒐集鋪面資料，過去檢驗技術大都為人力檢驗較費時且費工，故本研究嘗試以自動化鋪面破壞檢測系統進行路面檢測。
自動化鋪面破壞影像檢測系統(Auto Pavement Damage Image Detection System, APDIDS)採用雙線掃描相機進行非接觸式的鋪面檢測，而近年來影像處理技術發展相當迅速，不僅在相機等級的提升或者是影像辨識方法的提升，因此本研究嘗試使用兩台線掃描相機及相關影像擷取設備進行鋪面影像資料檢測，並開發一套自動化鋪面破壞影像辨識系統(Auto Pavement Distress Image Recognition System, APDIRS)，將自動化鋪面破壞影像檢測系統所擷取之影像進行辨識，其中包含辨識縱、橫向裂縫、坑洞、補綻、鱷魚狀裂縫、貓眼、人手孔等等鋪面常見破壞及附屬設施，並將破壞辨識結果進行鋪面狀況指標計算。
在自動化鋪面破壞影像辨識完成後建立鋪面破壞維護管理系統(Pavement Distress Maintenance Management System, PDMMS)，在鋪面破壞維護管理系統中建立道路基本資料、道路破壞位置資料及道路破壞統計資料以提供道路養護單位更有效率的管理整體道路之狀況。
研究結果發現使用雙線掃描相機所建置的自動化鋪面破壞影像檢測系統能有效的提升鋪面狀況調查之效率，並有效提升鋪面影像擷取之影像品質，而自動化鋪面破壞影像辨識系統採用分群演算法及分類演算法能有效的將破壞影像進行破壞分群並分類出破壞狀況，並有效的將辨識結果進行鋪面狀況指標計算，使鋪面檢測。
Pavement Condition Index, or PCI, is an important index of pavement service level, it not only affects the comfort of driving, but also provides the suitable time for pavement construction and rehabilitation. In order to manage pavement effectively, it is needed to inspect and collect pavement condition data on a regular base. Most of past inspection techniques are time consuming and labor intensive. The objective of this research was to develop an Auto Pavement Damage Image Detection System, or APDIDS, which can provide an automatic inspection technique.
This research used two linear scanning cameras and related image collection equipment to inspect pavement conditions. Moreover, it was then to develop an Auto Pavement Distress Image Recognition System, or APDIRS, which can identify the the common pavement defacts from APDIDS including: longitudinal, transverse cracking, potholes, patching, alligator cracking, manholes, and so on. The APDIRS can automatically evaluate the pavement defacts and conclude the value of PCI.
The research results indicated that by using two line scanning camera, the pavement inspection process enhanced effectively, the pavement image quality also raised. The APDIRS using clustering techniques and classification techniques could classification the damage image and calculate pavement circumstances index effectively.
This study concludes that APDIDS can adequately increase the rate of distress detection and quality of detected images in pavement condition surveying
the APDIRS incorporated with K-means and J48 can sort out different distress types and integrate into calculating the PCI. A Pavement Distress Maintenance and Management System, or PDMMS, was established to incorporate with data from APDIRS which can provide the profile of the roadway network and chronicle data of distresses. Thus, the agencies can efficiently manage the pavement conditions among districts. |
