You are on page 1of 5

MORPHOLOGICAL OPERATIONS TO RESTORE DAMAGED IMAGES

ANNAPOORNA H S
PG Student, ECE Department, DBIT, Bangalore, Karnataka, India
annapoornahs143@gmail.com

DR.MAHESH P K
Prof. & HOD, ECE Department, DBIT, Bangalore, Karnataka, India
mahesh24pk@gmail.com

ABSTRACT: Images form a prime source of information. Images produced display and contain
lots of information. Degradation of image may occur due to improper focusing of camera lens ,
relative motion between camera and object, sensor noise, atmospheric turbulence etc. Image
restoration being very important can be defined as a means of repairing, rebuilding or
reconstructing a original image form damaged one. Image processing includes extraction of
features, suppression of noise , edge detection, texture analysis , image segmentation , restoration
and reconstruction of image etc. In this paper we have implemented six morphological operations
using OpenCV on Intel atom processor. The operations include dilation, erosion, closing, opening,
region filling and boundary extraction.

KEYWORDS: Image restoration , Morphology and OpenCV .

INTRODUCTION

Image restoration is one of the important and emanating field of image processing. Image
restoration means repairing, rebuilding or reconstructing a original image form damaged one. In
the degraded image the degradation can be known or unknown. The amount of image that can be
restored depends on how much knowledge we have regarding original image, the amount of image
degraded and which model we are using to restore image.

Morphology is a branch of biology which deals with study of structure and form of plants and
animals . Mathematical morphology deals with is a theory and technique for the analysis and
processing of geometrical structures, based on set theory, lattice theory, topology, and random
functions. MM is most commonly applied to digital images, but it can be employed as well on
graphs, surface meshes, solids, and many other spatial structures.

Human vision is a complex combination of physical, psychological and neurological processes


that allow us to interact with our environment. Humans use their vision to see, observe detect,
track ,understand the objects, to navigate and to create a map of surrounding environment. The
objective of computer vision is mainly to design computer systems that has the capability of
performing these tasks efficiently and accurately in real-time and also using limited resources.
Computer vision has its applications in surveillance and other fields where accurate and very fast
processing is demanded in videos and live stream.

Remote sensing community and computer vision community both share many of the same goals
but its applications is not only focused in characterising the earths surface but also includes a
much vast range of applications. Computer vision applications include biomedical, industrial,
scientic, surveillance, environment exploration, understanding of documents, analysis of video,
games ,graphics and entertainment.

OpenCV was introduced in 1999 , it is an open source computer vision library which contains
more than 500 algorithms for analysis of video and images, which also includes inspection of
factory, security, medical imaging, user interface, stereo vision, camera calibration and robotics. It
has been largely used by researchers, scientists and developers as primary development tool in
computer vision.

OpenCV version 1.0 was introduced in 2006 and OpenCV 2 was launched in 2009 a second
major release which proposed some important changes, especially the C++ interface.The library
contains C and C++ interface. Due to this new interface the number of lines in code are reduced
which in turn also reduces programming errors.

OpenCV version 2.2 library is divided into multiple modules, where each and every module is
understood, in general, as these modules are dedicated to one group of computer vision
problems.OpenCV was mainly designed to focus on realtime applications and provide efficiency
in computation . OpenCV takes advantage of multicore processors and its library is written in
optimized C.

Most researchers, practical programmers and computer scientists are aware of role of computer
vision ,its applications and its benefits. There may be few people who are aware of all the ways in
which computer vision is used. Computer vision has its application in surveillance and also used
for analyzing and processing images and videos.

LITERATURE SURVEY

Reginald L. Lagendijk and Jan Biemond [1] et al., Images contain enormous amount of
information .Images may be degraded due to numbers of reasons . Degradations may be due to
many reasons.Wide range of degradations exist they are geometric distortion, bur etc.

George Bebis, Dwight Egbert and Mubarak Shah [2] et al., Vision is one of the important sense
organ of humans. Computer vision, enables computers to understand ,analyze and interpret
information from video streams and images.

Amalorpavam.G, Harish Naik T, Jyoti Kumari and Suresha M [3] et al., Image processing has
wide range of applications .In image processing the images are used to represent morphological
operations using a tool known as mathematical morphology .The authors have implemented the
morphological operations in MATLAB.

Sangita S L, Dr.S.S.Kanade [4] et al., In image processing , image restoration is an important


process. It is a mainly process to reconstruct original image from distorted image. The restored
image as applicationsapplied in many application in vast range of areas. The authors have also
discussed about various sources of noise and proposed denoising algorithm.

75
METHODOLOGY

The basic operations of morphology are erosion and dilation.Erosion is mainly used to diminish
the image and eliminate boundaries. These operations are mainly applied to binary image. These
operations can be performed on grayscale images and RGB(Red green and blue) images but more
sophistication is required.

Erosion is used to diminish images , eliminate boundaries whereas dilation is used to expand
images. Erosion can be use in thinning of images, where dilation can be used in thickening of
images. Dilation fills the holes or lines in damaged image and helps in restoration of image
As mentioned erosion and dilation are two basic operations of morphology ,combining these two
can produce complex operation . Two such operations obtained by combining erosion and dilation
are opening and closing.

Opening is process of performing erosion operation followed by dilation operation using same
structuring element .Similarly closing is a process of performing dilation operation followed by
erosion operation using same structuring elements. Opening and closing operations can be
performed on binary images , grayscale images and RGB images.

Boundary extraction is also known Edge detection is considered as an important area in computer
vision , edge detection is important as the higher level processing depends on good sharp edges.
Gray scale images contain vast amount of data .Much of the data in this image is insignificant for
image processing operations. Hence some amount of unwanted data such as background of an
image is eliminated , so that significant objects are separated from background.

Region filling algorithm is based on dilation, intersection and complementation. In this the area
inside the boundary is filled in from start point .

EXPERIMENTAL RESULTS

The basic operations of Morphology that is erosion, dilation ,opening , closing , boundary
extraction and region filling is performed using open cv.

Figure 1. Input and Erode Image

76
Figure 2. Dilate and Opening Image

Figure 3. Closing and Boundary Extraction Image

Figure 4. Region Filling Image

77
CONCLUSION

In this paper six morphological operations including erosion, dilation, opening, closing, boundary
extraction and region filling are implemented in OpenCV version 2.0. Mathematical morphology
is an efficient and powerful tool to extract relevant features from images.

REFERENCES

Reginald L. Lagendijk and Jan Biemond (1999). Basic Methods For Image Restoration And
Identification.
George Bebis, Dwight Egbert and Mubarak Shah (2003). Review of Computer Vision Education
IEEE Transactions On Education, Vol. 46, No. 1
Amalorpavam.G, Harish Naik T, Jyoti Kumari and Suresha M (2013). Analysis of Digital Image
Using Morphlogical Operations .International Journal of Computer Science & Information
Technology.
Sangita S L , Dr.S.S.Kanade (2013). A brief review on image restoration in image processing,
International Journal of Advanced Research in Computer and Communication Engineering .

78

You might also like