A Hybrid Trademark Retrieval System Using Four-Gray-Level Zernike Moments & Image Compactness Indices

1. Department of Mechanical & Electro-Mechanical Engineering
Innchyn HER, Yung-Han HSIAO and Huan-Kai HUNG

2. CONTENT
 Introduction
 Objectives
 Trademark
 Methodology
 Smallest Enclosing Circle
 Compactness Indices
 Four Level Images
 Exclusive Trademark features
 Weighting and Normalization
 Results
 Conclusion
 Reviewer Comments

3. INTRODUCTION
 Still relies largely on manual and tag-based methods
 Using search codes risks the danger of inaccuracy, inconsistency,
and inefficiency.
 Content-Based Image Retrieval (CBIR)
 Contour-based approach
 Region-based techniques
 Zernike moments for trademark retrieval schemes

4. OBJECTIVES
 Propose a hybrid retrieval system for trademarks.
 Use Zernike moments for their resistance to noise, their invariance to
rotation, and multi-resolution capabilities.
 Two new image features are proposed; Four-Gray-Level and
Compactness Indices.
 In this system humans set up a benchmark and result of Hybrid
system will compare with benchmark.

5. TRADEMARKS
 A trademark is a mark that identifies one person's goods
 There are four different kinds of trademarks
(a)Word trademarks (b)Graph trademarks (c)Sign trademarks (d)Combined trademarks

6. METHODOLOGY
Smallest Enclosing Circle
 Zernike moments has property of rotational invariance. To achieve scaling
and translational invariance, additional arrangements are needed.
 Since Zernike polynomials are defined within the unit circle, a method for
conveniently finding the smallest enclosing circle for a trademark is
necessary.
 Propose a simplified version, based upon the Berg et al. recursive
formula, for locating the smallest enclosing circle for a trademark image.

7. SMALLEST ENCLOSING CIRCLE
 Firstly, extract digitally its contour, then, search
among the boundary points and find the two points
that are most distantly apart, say, points a and b.
Construct circle D1
Locate point c, which is farthest from the
center O1. Construct circle D2 by points a, b,
and c.

8. SMALLEST ENCLOSING CIRCLE
 Lastly, if there still are some remaining
points outside D2, there is a final
modification to make. Find point d that
is most distant from O2. Let
be the length of the new diameter,
construct D3, which is the smallest
enclosing circle used in this paper.
2 2O d O c

9. WRAP CONTOUR
 Besides Zernike moments, a region-based method we need some
contour-based features that can represent the gross shape.
 Propose a wrap contour concept.
(b) (c)
(a)
Using contours as features.
(a) A trademark and its smallest enclosing circle,
(b) The convex hull for that trademark,
(c) The wrap contour for that trademark.
Locate smallest enclosing circle.
Radial lines are drawn
The intersections are recorded as a
function r()
If there is no intersection, r=0
Finally, let the wrap contour r()
become a closed and continuous
curve in the polar plane

10. COMPACTNESS INDICES
 The wrap contour used to define two new features called image
compactness indices, CI1 and CI2. They can be regarded as mixed
contour-based and region-based features, and are defined as:
Compactness index 1 (CI1) =
Compactness index 2 (CI2) =
Area of Wrap Contour
Area of Smallest Enclosing Circle
Image Area
Area of Wrap Contour

11. FOUR-LEVEL IMAGES
 For simple images as trademarks, some researchers suggested using
only a two-level (binary) scale. Transforming trademarks into binary
images sometimes incurs significant losses in the features.
Losing boundaries from
using too few gray levels.
(a) Some internal boundaries
are missing,
(b) The plus sign is gone.

12. FOUR-LEVEL IMAGES
 We propose using four gray levels that correspond to levels 0, 85, 170,
and 255, in a 0 to 255 grayscale.
 A trademark image is then transformed, with evenly spaced threshold
values, to an image that contains only these four gray levels.
 By thus using more than two gray values, the loss of image features is
effectively reduced.

13. EXPERIMENT AND RESULTS
 In this paper, 2020 trademark images were collected and compiled to
form the experimental database.
 These Table used as benchmarks to tested hybrid scheme.
Human-Picked
Groups of Similar
Trademark
Patterns
Nike
Goblet
Recycle
Michelin
Monogram

14. EXPERIMENT AND RESULTS
 A collection of eight weighting factors
 w1: compare the whole input image with whole images from the
database.
 w2: compare core images only.
 w3: compare the core of the input image with the whole images from
the database. This is to remove the frame of the input image.
 w4: compare the inversed core of the input with the whole database
 w5: compare CI1's of whole images.
 w6: compare CI2's of whole images.
 Lastly, w7 and w8: compare CI1's and CI2's of the core images, respectively.
j 1 2 8
w {w , w , w } K

15. EXPERIMENT AND RESULTS
 Using different gray-levels with Zernike moments.
(a) Two-level scheme
(b) Four-level scheme
Input
image
Input
image
Gathers more monogram-like images in its top rows

16. EXPERIMENT AND RESULTS
(a) Hu's moment invariants
(b) Traditional Zernike moments
(c) Kim's modified
Zernike
(d) Our compactness
index CI1
w5=1 (compare CI1's of whole images)

17. EXPERIMENT AND RESULTS
By using more image features to improve the search efficiency
(a) Hu's moment invariants
(b) The original Zernike,
(c) Kim's modified Zernike

18. EXPERIMENT AND RESULTS
(d) Our four-level Zernike
(e) Our Compactness indices
(f) Combining double Zernike and Compactness indices
{1,4,0,0,0,0,0,0}
{0,0,0,0,1,0,4,1}
{1,4,0,0,1,0,4,1},

19. PERFORMANCE SUMMARY
 In average, both our double Zernike schemes and CI-only schemes got
roughly 60% of human-picked images in image listings.
 This was no better than Kim's modified Zernike (65% in average) and
more or less equivalent to traditional Zernike (58% in average).
 But Combination of double-Zernike and CI's yielded very good (95% in
average) results.
 A 64×64 image size was used for the input and an input image must go
through a series of preprocessing processes.
 All image features are then extracted, paired with their weighting factors
and compared with stored data in the database.

20. CONCLUSION
Both region-based image features (i.e., four-gray-level Zernike moments)
and contour-base features (the proposed compactness indices) were
used.
A simplified method for finding the smallest enclosing circle was also
presented.
Features of different categories were conjoined via a weighting scheme.
Experiments have verified that, when human viewpoints were used as
standards, this hybrid scheme performed considerably better than some
existing retrieval methods.

21. REVIEWER COMMENTS
 Weighting factors are not clear.
 Experiments needs better organization and more comparison. Result
comparison with table.
 Proposed method should compare with recent work. The literature
review is too limited to trademark retrieval (its an image processing
paper).
 What kind of shape can not be handled well? More justification
needed? With some images.
 Smallest enclosing circle algorithm is not complete. No proof of the
validity. Empirical method.

22. REFERENCES
 H. Freeman, "On the Encoding of Arbitrary Geometric Configurations," IRE Trans. Electronic Computing, vol.
10, pp. 260-268, 1961.
 H. L. Peng and S. Y. Chen, "Trademark Shape Recognition Using Closed Contours", Pattern Recognition
Letters, vol. 18, pp. 791-803, 1997.
 C. C. Huang and I. Her, "Homomorphic Graph Matching of Articulated Objects by An Integrated Recognition
Scheme," Expert Systems with Applications, vol. 31, pp. 116-129, 2006.
 P. Y. Yin and C. C. Yeh, "Content-based Retrieval from Trademark Databases," Pattern Recognition Letters, vol.
23, pp. 113-126, 2002.
 M. K. Hu, “Visual Pattern Recognition by Moment Invariants,” IRE Trans. Information Theory, pp. 179-187,
1962.
 G. Ciocca and R. Schettini, “Content-based Similarity Retrieval of Trademarks Using Relevance Feedback,”
Pattern Recognition, vol. 34, pp. 1639-1655, 2001.
 N. K. Kamila, S. Mahapatra and S. Nanda, "Invariance Image Analysis Using Modified Zernike Moments,"
Pattern Recognition Letters, vol. 26, pp. 747-753, 2005.

23. THANK YOU!