About this document
Automatic Pavement Crack Detection by ali fares is a document available to read on EtoBox.
The document presents CrackTree, an automatic method for detecting cracks in pavement images, addressing challenges such as low contrast and shadows. The method involves a geodesic shadow-removal algorithm, the creation of a crack probability map using tensor voting, and the identification of cracks through a graph model and minimum spanning tree. Experimental results demonstrate that CrackTree outperforms existing crack detection methods on a dataset of 206 pavement images.
- Author
- ali fares
- Language
- EN