鐵路運輸系統在國家發展中扮演著非常重要的角色，近幾年各國不斷追求列車速度及大幅提升運輸量的情況下，增加軌道與車輪之間的接觸頻率導致軌道系統加劇損壞，當鋼軌損害達相當程度時將增加鋼軌與列車彼此間的接觸面積，重則造成交通事故，輕則影響乘客舒適度問題及行駛過程的噪音影響路線周遭住戶的安寧，管理單位首先必須即時且準確的量測鋼軌磨耗量確保軌道系統營運的安全，正確掌握全線段軌道不整程度並將鋼軌損害控制在規範容許值內，制定相關養護規範並系統性的管理軌道系統。 目前國外的非接觸式鋼軌檢測技術已經發展到相當程度了，而國內各單位於現階段主要從國外購買成套檢測車或採取檢測人員拿磨耗儀徒步檢測的方式，前者必須仰賴國外廠商而後者的檢測效率不高，而近年來非接觸影像處理方面的研究發展已經相當成熟，本研究以非接觸雷射掃描技術為主要核心，搭配影像處理技術與Matlab程式語言進行編寫鋼軌磨耗檢測程式，透過影像將實驗所獲得的量測值與標準鋼軌斷面參考值，計算實驗值與標準斷面參考值的差量得到該斷面磨耗值，利用手持式鋼軌磨耗儀進行實驗驗證，判斷誤差並修正，以尋求更有效率之檢測方式，未來可以發展出快速而精確的軌道構件幾何構形的檢測系統，提升維修之效率。 ;The railway transportation system plays a very important role in the developed countries. In recent years, as increase the speed and frequency of trains, the rail system is damaged more quickly. When the damage of the rail head reaches a degree, the contact area between the head of rail and the train will increase. These results carries the passenger comfort ability, noise problems and even lead to traffic accidents. The railway maintenance group must immediately and accurately measure the rail wearing condition to ensure the safety of the rail system operation. Taiwan inspect the wear damage manually have purchase authorities non-contact inspection vehicles from foreign countries .However, the demand on automatic and fast inspection technique is urgent. This study integrates non-contact laser technology, image processing technology and Matlab programming language to develop a rail wear detection system. The experimental measurement of the rail head profile obtained through the image processing and the reference value of the standard rail section is used to calculate the amount of wearing. Using a hand-held rail abrasion instrument for experimental verification to judge errors, This study sought more efficient detection methods and wanted to develop a fast and accurate track damage detection system to enhance the effectiveness of the maintenance work
