Affiliations: [a] Department of Control, Instrumentation and Robot
Engineering, Chosun University, Gwangju, Korea | Université Laval, Canada | Tohoku University, Japan
Correspondence:
[*]
Corresponding author: Jinyi Lee, Department of Control,
Instrumentation and Robot Engineering, Chosun University, Gwangju 501-759,
Korea. Fax: +82 62 230 7101; E-mail: [email protected]
Abstract: Inclusions in steel are entrapped foreign materials, which may be
metallic or non-metallic. The inclusions disrupt the structural homogeneity of
the steel and affect its mechanical properties. Moreover, nonmetallic
inclusions such as calcium and magnesium can be propagated as narrow cracks in
steel strips after the rolling process. In this paper, we propose the use of
magnetic flux leakage testing involving the fusion of images obtained by arrays
of linearly integrated Hall sensors (LIHaS) and linearly integrated giant
magnetoresistance sensors (LIGiS) for inspecting inclusions in steel strips. In
the proposed method, the normal (Bz) and tangential (Bx) components of the
magnetic leakage field around an inclusion are detected by means of LIHaS and
LIGiS, respectively. The images of the magnetic leakage field were obtained at
different scanning (magnetization) directions relative to the rolling
direction. The Bz and Bx images of the inclusion are then combined by using
image fusion technique to enhance the detection of the inclusion. The
effectiveness of the proposed image fusion technique was experimentally
verified by using LIHaS and LIGiS images of an inclusion in a cold-rolled steel
strip.
