You are on page 1of 4

International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882

Volume 3, Issue 3, June 2014

Detection of Broken Blister using Canny and Rc-algorithm


Munish Kumar Dhiman1, Dr Rajat Gupta 2
1

M.Tech Final Year Student, Deptt. of ECE,MMEC, MMU, Mullana, Ambala, Haryana, India
2
Professor , Deptt of ECE, MMEC, MMU, Mullana, Ambala, Haryana, India

ABSTRACT
Medication is an important part of everyones life.
Peoples are affected by many diseases at some point or
other in life and most of these diseases can be cured only
with the help of medicine. So there should be some
quality control on medicines. In todays world large
amount of medicines are produced to meet the high
demand for these medicines. During the production of
these medicines there may be damages like cracks,
breakage present in the tablets or capsules. Consumption
of these damaged tablets may cause some problem in
skin, eyes and face. Most of the tablets are not advisable
to be consumed in broken form. So these medicines
should be checked for any kind of damage before being
sold two the consumers. Since huge amount of
medicines are produced every day so manual inspection
of these medicines is very challenging and time
consuming task. Therefore a system is required to make
this process automated. This paper presents an approach
for automatic inspection of broken pharmaceutical drugs.
The approach is based on the use of edge detection
methods i.e. canny edge detection method and RCalgorithm to check for the defects in the tablets. And
give matched percentage of different pharmaceutical
drugs Blisters.

technology, image and signal processing for the purpose


of quality assurance/productivity enhancement in realtime.

II. PROPOSED SYSTEM ARCHITECTURE


Figure shows the architecture of the proposed system for
the detection of broken pharmaceutical drugs. It consists
of preprocessing, edge detection, and template matching
and calculating the matching percentage. The image is
captured and given as input to the system. In
preprocessing the image is processed in predefined
manner to make it suitable for further processing.
Preprocessing steps includes conversion to gray image
and enhancement.

Keyword -Tablets, Capsules, Image processing, Canny


Edge Detection, RC -Algorithm, Automated Visual
Inspection

I. INTRODUCTION
Digital image processing involves processing of digital
images by using various image processing techniques
algorithms. Digital image processing has its role to play
in science and technology. Various image processing
techniques such as feature extraction, template matching
and pattern recognition, and edge detection to process
digital images can be very advantageous in many
applications. With every passing day, the use of
automated system is increasing. This is particularly
relevant if it is necessary to acquire more exact
information or if information is needed faster than a
person is capable of providing it or if we want to replace
a human task with a technical application. It has become
possible due to advancement in the field of sensor
www.ijsret.org

Figure 1: Architecture of Proposed System

643

International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882
Volume 3, Issue 3, June 2014

Edge detection in digital image marks the regions where


the brightness changes sharply. Image enhancement
involves removal of noise from the image using
thresholding. Conversion of gray scale to binary image
is called thresholding [1]. Then the image is filtered
using the low pass filter operator called Canny Edge
detector. Canny edge detection operator is used to obtain
each tablets boundary in a blister [2] [3] [4]. A template
image is saved in the system. The template image is the
image without defects. The template image undergoes all
the process similar to the test image. Then the test image
is compared to the template image. The pixels are
calculated for the template image and the input image.
The pixels are compared, if the template image pixel is
greater than the input image pixel then the there is a
defect in the blister. The user is provided with the output
console with the alert message and the number of pixels.

Figure 2 : Gray Image

III. ALGORITHM FOR IDENTIFYING


DEFECTS IN PHARMACEUTICAL DRUGS
Step 1: Read the Blister as input image.
Step 2: Display the input image.
Step 3: Convert the input image into grayscale image.
Step 4: Display the converted grayscale image.
Step 5: Add random noise to image.
Step 6: Apply ACO for heuristic search of edges.
Step 7: Apply canny and Rc- algorithm for Edges
detection.
Step 8: Got 6 edge maps from Edge detection by using
thresholding.
Step9: The image of the Blister without any damage is
taken as a template image. The template
image
undergoes all the pre-processing steps similar to the
input image.
Step 10: Display the template image.
Step 11: Compare the input image and the template
image.
Step 12: Display the compared images.
Step 13: Display the total matching result of two
different Blisters.

Figure 3: Image with Random Noise

Figure 4: Image With Compared Blisters

www.ijsret.org

644

International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882
Volume 3, Issue 3, June 2014

single tablet which is compared to each tablet in the


blister. If the shape varies then the package is identified
as defective tablet. The decision is made based on
number of matched pixels in two images. The proposed
system was able to efficiently detect the broken
pharmaceutical capsules and tablets when tested on
software.

V. CONCLUSION

Figure 5: After Applying Rc And Canny Detection

The manufacturing of tablets and capsules requires a


quality inspection system that not only keeps up with the
high production throughput, but also performs accurately
and reliably. The undesirable effects present in
pharmaceutical drugs can be unsafe from consumer as
well as producer point of view. So, its necessary to
detect and classify them. We used image processing
tools to detect these pharmaceutical defects. The
Proposed work approach covers all the aspects of defects
related to shape, size of the pharmaceutical drugs. The
algorithm can be implemented in various digital image
processing environments and can be part of complex
automated manufacturing and testing system.

REFRENCES

Figure 6: Image with Canny and Rc-algorithm apply on


template blister

IV. RESULTS AND DISCUSSION


The proposed algorithm was implemented with different
types of sample of capsules and tablets. One of the tablet
samples undergoes the steps to find whether there is any
defect in the blister. In the process, the input image is
converted into gray scale image (Figure 2). Random
noise is added to the gray scale image (Figure 3) which
makes the image fit for further processing. Canny edge
detection operator and rc- algorithm (Figure 5) is used
for edge detection to filter the tablets from its
background. The template blister image is stored in the
system based on the type of input blisters. The template
blister image undergoes all the pre-processing steps
similar to the test image. The template blister image is of

[1] R.Thilepa A Paper On Automatic Fabrics Fault


Processing Using Image Processing Technique in
Matlab, Signal & Image Processing: An International
Journal (SIPIJ) Vol.1, No.2, December 2010.
[2] Dawei Qi, Yuanxiang Li and Lei Yu The
Application
of
Mathematical
Morphological
Optimization Algorithm in Edge Detection of Defected
Wood Image , Proceedings of the IEEE International
Conference on Automation and Logistics Qingdao,
China September
2008.
[3] Peng Zhao and Shutao Li Tablets Vision Inspection
Approach Using Fourier Descriptors and Support Vector
Machines The 9th International Conference for Young
Computer Scientists,2008.
[4] Joze Derganc, Bostjan Likar, Rok Bernard, Dejan
Tomazevic, Franjo Pernus,Real-time automated visual
inspection of color tablets in pharmaceutical blisters,
Department of Electrical Engineering, University of
Ljubljana, May 2002.
[5] Ramya. S, Suchitra. J and Nadesh R.K ,Detection of
Broken Pharmaceutical drugs using Enhanced Feature
Extraction Technique,vol 5,no 2,2013.
[6] Hardeep Kaur and Er. Nidhi Garg ,Inspection of
Defective Pharmaceutical Capsules using Harris
Algorithm, International Journal of Advances in
Electronics Engineering, 2009.

www.ijsret.org

645

International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882
Volume 3, Issue 3, June 2014

[7] tefan Oprea, Ioan Li, Mariana Jurianu, Daniel


Alexandru Vian and Ion Bogdan Cioc ,Digital Image
Processing Applied in Drugs Industry for Detection of
Broken Aspirin Tablets, IEEE,ISBN: 978-1-4244-39744, 2008.
[8] Yahia s. al-halabi, Hesham jondi abd. new waveletbased techniques for edge detection, Journal of
Theoretical and Applied Information Technology,pp 3138,2010.
[9] Abha Sharma, Sugandha Arora,Inspection and
Classification of Defects in Pharmaceutical Capsules

Using Neural Network, International Journal of


Engineering Research and Development, Volume 1,
Issue 10, June 2012.
[10] Young-Beom Lee, Unsang Park and Anil K. Jain,
PILL-ID: Matching and Retrieval of Drug Pill Imprint
Images, International Conference on Pattern
Recognition, 2010.
[11] Mohammed J. Islam, Saleh Basalamah, Majid
Ahmadi and Maher A. Sid-Ahmed,Capsule Image
Segmentation in Pharmaceutical Applications Using
Edge-Based Techniques,

www.ijsret.org

646

You might also like