Page 74 - Riem-Vol6_nº3

419 IBRACON Structures and Materials Journal • 2013 • vol. 6 • nº 3
A. P. Martins | J. C. Pizolato Junior | V. L. Belini
Where N and M represent the quantity of rows and columns on
the image, and Q represents the quantity of interest pixels on the
image.
Finally, the block 6 in Figure 5 calculates the percentual of the
region occupied by crack on the input image. This percentual is
obtained at the comparison between the quantity of crack pixels
and the total pixels quantity on the image. This pixels compari-
son is accomplished independent of the opening, length and index
(quantity) of the cracks found on the input image.
The processing algorithm output information, which corresponds to
the percentual of crack occupation of the structure on masonry, is
exhibited directly on the mobile phone’s screen.
5. Results and discussions
The performance and efficiency evaluation of the proposed meth-
od was conducted by means of comparisons between measuring
recorded by the mobile phone and those from the conventional
method, on different crack openings. The comparisons between
the proposed and the conventional method were accomplished on
concrete specimen with standard shape, on a compression test
machine. The proposed method may be applied on masonry and
concrete structures. The objective of the experiment with speci-
men and testing machine was to obtain cracks for dimensional
analysis in a faster way, in comparison to the collect on periodic
inspections on masonry.
Different types of specimen composition may be used, once the
experiments do not take into account the mechanical characteris-
tics of the specimen. Figure 7 illustrates a specimen in the com-
pression test machine.
The experiment with the testing machine was carried out on six
cylindrical specimens, with a diameter of 100 mm and length of
200 mm with perpendicular faces. The surface monitored with the
proposed system (masonry or concrete) was prepared by adding
white acrylic paint on the specimen surface in order to facilitate the
subsequent crack observation. In the experiment were recovered
three specimens that presented bigger cracks focused on the fron-
tal region of the testing machine. The specimens were chosen aim-
ing to facilitate the measuring by the conventional method and the
positioning of the acquisition system, for the image capture on the
proposed method. After the breaking beginning of the specimen,
an advancement of compression’s axis of the testing machine, was
applied gradually, which promoted proportional and observables
crack increases on the specimen.
The conventional measuring of the crack opening was performed
by two technicians, using initially a 0.5 mm resolution graduated
scale. Due to the low scale resolution, other conventional mea-
suring was accomplished utilizing a 0.01 mm resolution pachym-
eter. On the comparison experiment were accomplished in each
method (proposed and conventional) four measures, resulting in
22 average measures to be presented in this section. The techni-
cians that performed the monitoring in a conventional method also
used the proposed system on the same conditions and specimens.
Figure 6 – (a). Image of crack on masonry represented in Gray-level
with 256 levels of grey. 6(b). Image after binarization process
A B
Figure 7 – Testing machine used in the civil
engineering to verify behavior and r
esistance to compression specimen