361
ISSN 1061-8309, Russian Journal of Nondestructive Testing, 2020, Vol. 56, No. 4, pp. 361–368. © Pleiades Publishing, Ltd., 2020.
Inspection and Monitoring
of Concrete Structures via Radiography
and Weighted Nuclear Norm Minimization Method
Amir Movafeghi
a
, Effat Yahaghi
b
, Behrouz Rokrok
a
, and Mahdi Mirzapour
c,
*
a
Reactor and Radiation Safety Research School, Nuclear Science and Technology Research Institute (NSTRI),
Tehran, Iran
b
Department of Physics, Imam Khomeini International University, Qazvin, Iran
c
Department of Mathematics, Faculty of Basic Sciences, Bu-Ali Sina University, Hamedan, Iran
*e-mail: m.mirzapour@basu.ac.ir
Received May 11, 2019; revised July 24, 2019; accepted July 26, 2019
Abstract—The location of reinforcement bar in concrete, the bar corrosion, diameter and the depth
below the surface are important factors in the evaluation of the load bearing capacity and the usage
property within the scope of building condition and damage analysis. Radiographic techniques are
used to examine the structure of different buildings. The basic goal of the inspection is the visualiza-
tion of reinforcement bars, fittings or tension cables. In this research, the Weighted Nuclear Norm
Minimization method was used to improve visualization of the hidden structures and defects from the
concrete radiographs. The method relies on minimization of the image energy and singular value
decomposition for enhancing contrast. The proposed algorithm was successfully applied to radio-
graphic images of concrete segments. Improvement of the structure detail visualization and defect
region detection were achieved while preserving structure edge and fine detail imaging information.
Evaluation of the image quality enhancement showed that the contrast level to noise increases by a fac-
tor of about two in reconstructed images using the proposed method. Also, the background is sup-
pressed towards the zero baselines, and better visual quality is achieved.
Keywords: concrete inspection, industrial radiography, non-destructive testing, weighted nuclear
norm minimization method, condition monitoring
DOI: 10.1134/S1061830920040087
1. INTRODUCTION
Post-construction testing of concrete structures is necessary to ensure the structure continues to meet
its design criteria. Concrete properties including the density, elastic modulus, strength, surface hardness
and surface absorption, and reinforcement position, size and distance from the surface can be measured
using different destructive and non-destructive testing methods [1–3]. In many destructive inspections
such as core testing and pull-out pull-off testing, the surface ought to be repaired after the test. Some tests
need to be performed non-destructively; non-destructive testing such as ultrasonic and radiography can
be applied to concrete structures for quality control or the resolution of doubts about the quality of the
materials or the construction itself.
The position of the reinforcement steel bars in concrete, the bar diameter, corrosion, and also the depth
below the surface can be estimated by the radiography. The same method can also reveal any flaws and for-
eign objects, the presence or absence of grouting in post-tensioned construction and variations in the con-
crete density [1, 4–6]. However, the radiography image quality may be significantly reduced by scattered
X-rays requiring the use of some form of scattered photon absorption/collimation to achieve reasonable
image quality. Although techniques such as photon collimation, masking, filters, back shielding by lead and
using different types of intensifying screens can be beneficial, they inevitably lead to higher radiography cost.
Image processing methods are by comparison significantly cheaper to implement and can aid interpretation
of concrete radiographs by improving the imaging contrast and quality [1, 2]. Simple to apply digital image
processing methods can now be offered to the operator that may, in the background, be sophisticated and
composite computational algorithms running on standard personal computers. The aim of the majority of
such methods is improving image detail visualization by reducing the background blurriness. Digital image
RADIATION METHODS