Identify Defects in Gears Using Digital Image Processing

International Journal of Engineering Research and Development
ISSN: 2278-067X, Volume 1, Issue 6 (June 2012), PP.49-55
www.ijerd.com

Identify Defects in Gears Using Digital Image Processing
Amandeep Mavi1, Mandeep Kaur2
1
Mtech Student of Dept.of Computer Science Engineering, SVIET-Banur, India.
2
Lecturer of Dept.of Computer Science, SVIET-Banur, India.

Abstract—Gear defects are a major reason for poor quality and of embarrassment for manufacturers. 0Inspection
processes done on these industries are mostly manual and time consuming. To reduce error on identifying gear defects
requires more automotive and accurate inspection process. Considering this lacking, this research implements a Gear
Defect Recognizer which uses computer vision methodology with the combination of local thresholding to identify
possible defects. The recognizer identifies the gear defects within economical cost and produces less error prone
inspection system in real time. In order to generate data set, primarily the recognizer captures digital gear images by
image acquisition device and converts the RGB images into binary images by restoration process and local threshold
techniques. Later, the outputs of the processed image are the area of the faulty portion and compute the possible defective
and non –defective gear as an output.

Keywords—Defect detection, image processing, computer vision, thresholding, counting number of teeth’s.

I. INTRODUCTION
All plastic industries aim to produce various competitive plastic products. The competition enhancement
depends mainly on productivity and quality of the plastic produced by each industry. In this sector, there have been an
enlarge amount of losses due to defective products. Most defects arising in the production process are still detected by
human inspection. The work of inspectors is very tedious and time consuming. The identification rate is about 70%. In
addition, the effectiveness of visual inspection decreases quickly with fatigue. Digital image processing techniques have
been increasingly applied to plastic gear samples for analyzing the product.

As the technological progress is [1] happening the products are now extensively made using plastic material
especially in robotics which needs to be ultra light weight and modular in nature plastic components like gears, As per
industry statistics we have found that gears are made up of plastic material High-density polyethylene (HDPE) [1] which
is prone to various kinds of defects when manufacturing using image processing. Therefore we suggest a fully robust
system taking advantage of image processing techniques (Image segmentation, Non smooth corner detection etc) must be
explored to build an economical solution to provide Total Quality Management[2] in manufacturing units which would
allow an eco-system of continuous monitoring and improvement there by reducing the cost.

This paper is organized into Section I includes Introduction, Section II Related work, Section III Model
Presentation, Section IV Results and Section V Conclusion and future work.

II. RELATED WORK DONE
Alisha Tremaine et.al [3] were able to understand how mould and die when not working properly, might give
rise to the surface defects in plastic product. This paper basically discuss defects develop due to shrinkage, due to
overheating and variation in temperature. To identify these kinds of defects, this paper suggests using SEM (Scanning
Electron Microscope) technique. This algorithm was running on the surface as well as the cross-sections of the plastic
products. The main objective of this paper was to perform a systematic study of several types of internal defects that can
be occasionally found in forged products and identify the defect surface prior to and after high temperature heat
treatment.

S. Kamaruddin et.al [4] paper presented a study in which an attempt has been made to improve the quality
characteristic (shrinkage) of an injection molding product (plastic tray) made from blends plastic (75% polypropylene
(PP) and 25% low density polyethylene (LDPE)) by optimizing the injection molding parameters using the Taguchi
method. This paper had made an attempt to describe the optimization of the injection moulding process parameters for
optimum shrinkage performance of a plastic tray which is made from polymer blends or polyblends.The performance of
the plastic trays was evaluated in terms of its shrinkage behavior. The analysis of the results shows that the optimal
combinations for low shrinkage are low melting temperature, high injection pressure, low holding pressure, long holding
time and long cooling time. Using Taguchi method for design of experiment (DOE), other significant effects such as
interaction among injection molding parameters are also investigated.

Bernd Scholz-Reiter et.al [5] introduced a machine vision based quality concept for the surface inspection of
micro parts using these kinds of three dimensional images, called height maps. Quality control, the use of flexible and
robust machine vision techniques are necessary because the exact positioning of micro-parts was not always given and
has to be virtually compensated. By obtaining high-resolution optical images with depth selectivity, it allows three-
dimensional reconstructions of topologically-complex objects. This paper outlines a machine vision based quality
49


concept for the surface inspection of micro-parts being manufactured in the Collaborative Research Center (CRC).To
detect manufacturing errors on the surface, the neighborhood of each pixel is scanned and the deviation to the arithmetic
mean of the pixel values are computed which correspond to neighborhood area. To detect manufacturing errors on the
surface, the neighborhood of each pixel is scanned and the deviation to the arithmetic mean of the pixel values from the
corresponding neighborhood area is computed.

III. MODEL PRESENTATION
The system design of gear defect recognizer, which mentioned into this paper, is illustrated in Fig. 1. The
proposed system can be a competitive model for recognizing gear defects in real world. Base on the research, the
proposed system design is separated into two parts. The first part of our research processes the images to calculate the
thresholding values of different gears. The second part calculates the number of gears and third part checks whether the
gear is defective or non-defective.

A. Processing Input Using Computer vision methodology:
In our recognition system, the original digital (RGB) image is converted into grayscale image through noise
removing and filtering techniques (restoration process). As image processing is costly, for this reason, adaptive median
filter algorithm has been used as spatial filtering for minimizing time complexity and maximizing performance. After
restoration processing, we calculate threshold value of grayscale image. In our proposed system, the most important key
point is the decision tree processing in order to achieve the threshold value. As we know that there have been different
types of color gear images and also different types of defects in gears, so local thresholding was used based on decision
tree process. We have identified the threshold value (T) at greater than 120 and less than 60. Due to different threshold
values to different pattern of faults of gears, we generalize a specific threshold value (t) for all types of gears. Base on the
threshold value achieved from the decision tree, grayscale image is converted into binary image using local thresholding
technique.

Figure 1. System Design of Textile Defect Recognizer

50


B. Teeth counting algorithm of gear

Figure 2. Flow chart for Calculating teeth’s of plastic gear

The steps for teeth calculating algorithm of plastic gear is as follows:-
1. The input image is read by using imread function.
2. The algorithm can be tested for gray scale and a color image by appropriately using functions such as is gray
function.
3. Denoise the gear image for removing the small dirt particles.
4. Find the connected components of the gear image .Pixels are connected if their edges touch. This means that a
pair of adjoining pixels are part of the same object only if they are both on and are connected along the
horizontal or vertical direction..
5. Generate the labeled matrix and produce an RGB image for highlighting the gear image.
6. Extract regional properties of the gear image.
7. Erode image.
8. Results.

C. Algorithm for purposed work

Figure 3. Flow chart of the purposed method

51


The proposed algorithm is mentioned as follows:-
1. The input image is read by using imread function.
2. The algorithm can be tested for gray scale by appropriately using functions.
3. Enhancement image quality by using the spatial filters which operate on pixel values.
4. Denoising -Noise is the result of errors in the image acquisition process that result in pixel values that do not
reflect the true intensities of the real scene.
5. Segment largest object i.e. gear image.
6. Calculate the labels i.e. number of tenths of gear.
7. If the number of teeth match with the subscribe number then the gear is non-defective, otherwise the gear is
defected.

IV. RESULTS
To see the qualitatively as well as quantitatively performance of the proposed algorithm, some experiments are
conducted on several colored and gray scale images. The effectiveness of the approach has been justified using different
images. The results are computed qualitatively (visually) as well as quantitatively using quality measures.

The figures from Figure 4 to Figure 13 are the screenshots of the proposed work which shows the different
images which consists of original images and output gear images.

Figure 4.Original gear image

This is the RGB image of the original plastic gear which is used as an input.

Figure 5.Segmentation process

Here the number of the gray values in the value was 255 which have been reduced to 35 with segmentation process.

Figure 6.Grey scale image

The gear images have been converted into black by using the complement code for increasing the visibility.

Figure 7.Binary image

The above figure image shows the binary image for increasing the visibility with respect to the surface.

52


Figure 8.RGB image

This is the colored image of a gear for highlighting the gear part.

Figure 9.Input image

This is the binarized image of the gear passing through the Teeth counting algorithm.

Figure 10.Circumference of gear

The above image shows the step of a teeth calculating algorithm where it identify where it identify the outer as
well as inner circumference of teeth based on Teeth counting algorithm.

Figure 11.Count number of teeth’s

The above image count the number of teeth’s of the gear image for checking whether the number of teeth’s are
same as that of the subscribe number. If the number of teeth matches with the subscribe number then the gear in non-
defective otherwise it is defected.

53


Figure12.Original image

This figure shows an RGB image of the original plastic gear which is used to make industrial products like robotics.

Figure13.Defected gear

The above image basically shows the location of the defect after running the defect identifier algorithm based
on thresholding.

V. CONCLUSION AND FUTURE SCOPE
Firstly, in this paper we have been able to detect defective gears which have specific number of teeths.If the
number of teeth’s are deformed or more or less than the subscribe number than the gear is defective in nature and
therefore the gear is useless. Secondly, we have been able to identify some surface defects by using range bound
thresholding which can work on both color and grey scale images of the gears.

After conducting the above procedure for defect detection we suggest that in future we must take advantage of
some machine learning algorithms for making defects detecting more reliable and robust. There are number of future
possibilities for improving the performance of these detection algorithms like usage of machine algorithms which help to
identify the defective parts as these occur over a period of time. They increase their accuracy based on the updated
parameter set and scenario machine algorithm like Support Vector Machine, K-NN and neural network can be used.

VI. ACKNOWLEDGMENT
I would like to thanks my parents and my friends for their support and trust.

REFERENCES
[1]. Amandeep Mavi, Mandeep Kaur,(2012) ―Identify defects in Plastic (gears) using Digital image processing -A Review
IRACST - International Journal of Computer Science and Information Technology & Security (IJCSITS), ISSN: 2249-9555
Vol. 2, No.2, April 2012
[2]. http://searchcio.techtarget.com/definition/Total-Quality-Management.
[3]. Alisha Tremaine (2005), ―Characterization of Internal Defects in Open Die Forgings‖ FIERF Grant Project for
Undergraduate Research November 17, 2005.
[4]. S. Kamaruddin, Zahid A. Khan and S. H. Foong(2010) ― Application of Taguchi Method in the Optimization of Injection
Moulding Parameters for Manufacturing Products from Plastic Blend ‖ IACSIT International Journal of Engineering and
Technology, Vol.2, No.6, December 2010 ISSN: 1793-8236.
[5]. Bernd Scholz-Reiter, Michael Lütjen, Hendrik Thamer, Dennis Dickmann, ―Towards Machine Vision based Surface
Inspection of Micro-Parts‖, Recent Advances in Applied and Theoretical Mechanics, ISSN: 1790-2769 ISBN: 978-960-474-
1403.

54


Authors Profile

Miss. Amandeep Mavi was born in the small village of Mohali (Punjab).After finishing high school, she
move to Kharar to pursue Bachlor’s dergree in Computer Science. After graduating with a bachelor of Technology in
computer science from Guru Gobind Singh College of Modern Technology, Kharar. in 2009.Now she is pursuing
M.Tech in computer science engineering from Swami Vivekanand Institute of Engineering and
Technology,Banur,Punjab.

Mrs. Mandeep Kaur-She is Masters of Technology in Computer Science and Engineering from Punjab
Technical University with specialization in Networking and Digital Image Processing. She has given guidance to many
student in their thesis work of M.Tech.She has also contributed in the research work on WiMax networks with her
papers. She has four years teaching Experience and presently working as Lecturer in Swami Vivekanand Institute of
Engineering and Technology. She has done Bachelor's of Engineering from Punjab Technical University in the year 2005
in Computer Science and Engineering with Information Technology.

55

Identify Defects in Gears Using Digital Image Processing

Related slideshows

More Related Content

Identify Defects in Gears Using Digital Image Processing