Copyright © 2018 Authors. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted
use, distribution, and reproduction in any medium, provided the original work is properly cited.
International Journal of Engineering & Technology, 7 (4.5) (2018) 355-359
International Journal of Engineering & Technology
Website: www.sciencepubco.com/index.php/IJET
Research paper
Shape based Object Retrieval Technique for Vehicular
Spare Parts
Ravi Hosur
1,*
, Dr. Dayanand G Savakar
2
and Shilpa Madabhavi
3
1
Assistant Professor, Computer Science & Engineering Department, BLDEA’s V.P.Dr.P.G.HALAKATTI College of
Engg. & Tech., Vijayapur, Karnataka(INDIA)
2
Professor, Department of Computer Science, Dr. P. G. Halakatti Post Graduate Centre, "Vachana Sangama", Rani
Channamma University-Toravi,
Vijayapur, Karnataka(INDIA)
3
MCA Department, BLDEA's V.P.Dr.P.G.HALAKATTI College of Engg. & TechVijayapur, Karnataka(INDIA)
*Corresponding author E-mail:mca.hosur@bldeacet.ac.in
Abstract
Shape based object retrieval technique is an application used for the automatic retrieval of vehicular spare parts based on the input image,
this application can be useful in the automobile industry/shop for the retrieval of specific object from the group of objects. For the re-
trieval process first the database has been designed to store different shapes of nuts that are hexagon and square. With the application of
median filter and wiener filters in the preprocessing step, the canny edge-detector for identifying the edges and to ascertain shape of the
input and database images the convex hull approach is applied. This system is successful in retrieving the shape of the images which
already exist in the database and the method results with the accuracy of 82.1875%
Keywords: Shape, objects, Vehicular parts, retrieval, automobile, edge.
1. Introduction
The shape-based object retrieval system is an application used for
automatic retrieval of vehicular spare parts. The automatic retriev-
al of the vehicular spare parts in the automobile industry is one of
the important needs, because searching of the specific part from
the group of the objects consumes more time and energy. The
smart system in the automobile industry will be used to separate
the different spare parts of the vehicle. Instead of manual search-
ing we can make the automatic search by using this application
this reduces the human effort and saves time. We can specify the
shape based object retrieval technique as the system for fetching
the images from the database of the automobile industry/shop, the
database consisting of the different images of the vehicular spare
parts of different shapes of nuts such as hexagon and square. The
earlier image retrieval techniques use the text for the retrieval
process, in the text based approach the search is done through
keywords which will not provide accurate results. In the later days
content based approach has been developed, which work on the
observable characters of the image and the characters are color,
shape and texture. Among all the three contents important charac-
ter for the retrieval process is shape, hence in the proposed work
the shape is considered as the one of the important feature for the
retrieval of the vehicular objects. For the preprocessing step medi-
an filter or wiener filter is used depending on the noise present in
the image, then edge based segmentation is performed by using
canny-edge detection algorithm. As compared to other edge iden-
tification techniques, this approach provides better edges including
both strong and weak edges. After applying the canny edge mech-
anism it provides the distinct edges which are already exist in the
image and then by applying convex hull method the shape is ob-
tained from the image. Images which already exist in the database
are retrieved when shape of the input and database image are
matched.
The automatic retrieval of the vehicular spare parts in the automo-
bile industry/shop is one of the important needs. Manual searching
is the drawback so to overcome that, this application has been
designed by considering one of the important content of the im-
age, which is shape. The retrieval process is carried out by consid-
ering the shape, block diagram and algorithm discussed above
show the working procedure of the proposed work
2. Literature Survey
[1] Image retrieval can be performed by using different contents of
the image like color, texture and shape are used. By observing
these contents shape is one of the important content for retrieving
the images. In the proposed method image retrieval is done by
applying 2-D Fourier transform the Fourier descriptor (FD) is used
on the polar shape image.
[2] The technique which works on the content of the image treats
shape as one of the feature among the other features which are
used for retrieval. In the proposed method two important shape
descriptors are used for retrieving the shape of the image, the de-
scriptors are region based and contour based. By using these pro-
cedures retrieval procedure is performed.
[3] In the different types of applications like remote sensing, crime
prevention, for searching browsing and retrieving the images the
different devices are required. Because of this the various ap-
proaches are introduced. In the earlier days the text based ap-