Tracking and detection of cracks using minimal path techniques

Pavement crack minimal path methods

Kaul, Vivek ; Electrical and Computer Engineering

University:Georgia Institute of Technology

Department:Electrical and Computer Engineering

关键词: Pavement crack minimal path methods;   

Others  :  https://smartech.gatech.edu/bitstream/1853/37214/1/kaul_vivek_201012_phd.pdf

美国|英语

来源: SMARTech Repository

【 摘 要 】

The research in the thesis investigates the use of minimal path techniques to track anddetect cracks, modeled as curves, in critical infrastructure like pavements and bridges. Wedeveloped a novel minimal path algorithm to detect curves with complex topology that mayhave both closed cycles and open sections using an arbitrary point on the curve as the soleinput. Specically, we applied the novel algorithm to three problems: semi-automatic crackdetection, detection of continuous cracks for crack sealing applications and detection of crackgrowth in structures like bridges. The current state of the art minimal path techniques onlywork with prior knowledge of either both terminal points or one terminal point plus totallength of the curve. For curves with multiple branches, all terminal points need to be known.Therefore, we developed a new algorithm that detects curves and relaxes the necessary userinput to one arbitrary point on the curve. The document presents the systematic developmentof this algorithm in three stages. First, an algorithm that can detect open curves withbranches was formulated. Then this algorithm was modied to detect curves that also haveclosed cycles. Finally, a robust curve detection algorithm was devised that can increase theaccuracy of curve detection. The algorithm was applied to crack images and the results ofcrack detection were validated against the ground truth. In addition, the algorithm was alsoused to detect features like catheter tube and optical nerves in medical images. The resultsdemonstrate that the algorithm is able to accurately detect objects that can be modeled asopen curves.