Search

Image binarization is an important processing technology. This paper presents a new image binarization method based on dynamic gradient and global threshold. Firstly, a binary image A is generated by using the Otsu's method. Then, a binary image B is gotten with the dynamic gradient method. Lastly, the final binarization image C results from the logic AND operation using A and B. The experimental result shows that the method presented in this paper can preserve details and edges effectively compared with other binarization methods.

AbstractStructure safety assessment is one of the fundamental objectives for bridge health monitoring. In order to identify whether a bridge is safe, a bridge health model identification method based on Principle Component Analysis (PCA) is presented in this paper. A matrix made up of vazious measurement data of a bridge is constituted, the method of determine the similazity of bridges using the eigenvalue distance between data matrixes of bridges is addressed, and some issues related to the collection of the measurement data and the establishments of data model are also discussed.

AbstractIt is very important to evaluate the health status of a bridge by accessing cracks information of its key parts. Images of cracks aze obtained by analyzing the video information of the bridge’s key positions, the identifying algorithm and rebuilding algorithm are proposed based on the surface images of the bridge. Crack images pre-processing, crack images extracting, vector image storage and crack images rebuilding are addressed in this paper.