Abstract :

Image processing has been used for assessment of infrastructure surface coating conditions for years. In North America, civil engineers have utilized image recognition for steel bridge coating inspection since late 1990s. However, there is still no robust method to overcome the non-uniform illumination problem for infrastructure surface coating assessment to date.

In this seminar, a newly developed steel bridge coating assessment method, called the Adaptive Ellipse Approach (AEA), will be introduced. AEA adopts the a*b* color configuration of the L*a*b* color space as its backbone due to the fairly good light-filtering ability of a*b*. In AEA, a rust image is partitioned into three parts: background, rust, and mild�rust-color spots. The idea is to properly identify the mild-rust-color spots using an adaptive ellipse. Also, illumination adjustment, which is part of AEA, is used to overcome the non-uniform illumination problem. Finally, the performance of AEA is compared to the K-Means algorithm, one of the most popular and effective image recognition methods, to show the effectiveness of AEA.

Biography:

Dr. Po-Han Chen is currently an Assistant Professor in the School of Civil and Environmental Engineering, Nanyang Technological University, Singapore. He is specialized in project management, IT applications in construction, artificial intelligence, and image recognition. He got his Bachelor's degree from National Taiwan University, and Master's and Ph.D. degrees from Purdue University. He received the Best Paper Award in ISARC 2007 and is listed in various Who's Whos.