Efficient Computation of Zernike and Pseudo-Zernike Moments for Pattern Classification Applications 1

Size: px
Start display at page:

Download "Efficient Computation of Zernike and Pseudo-Zernike Moments for Pattern Classification Applications 1"

Transcription

1 REPRESENTATION, PROCESSING, ANALYSIS, AND UNDERSTANDING OF IMAGES Efficient Computation of Zernike and Pseudo-Zernike Moments for Pattern Classification Applications 1 G. A. Papakostas a, Y. S. Boutalis a, D. A. Karras b, and B. G. Mertzios a a Democritus University of Thrace, Department of Electrical and Computer Engineering, Xanthi, Hellas b Chalkis Institute of Technology, Automation Department Chalkida, Hellas (gpapakos, ybout, mertzios}@ee.duth.gr, dakarras@teihal.gr Abstract Two novel algorithms for the fast computation of the Zernike and Pseudo-Zernike moments are presented in this paper. The proposed algorithms are very useful, particularly in the case of using the computed moments, as discriminative features in pattern classification applications, where the computation of single moments of several orders is reuired. The derivation of the algorithms is based on the elimination of the factorial computations, by computing recursively the fractional terms of the orthogonal polynomials being used. The newly introduced algorithms are compared to the direct s, which are the only s that permit the computation of single moments of any order. The computational complexity of the proposed is O(p ) in multiplications, with p being the moment order, while the corresponding complexity of the direct is O(p 3 ). Appropriate experiments justify the superiority of the proposed recursive algorithms over the direct ones, establishing them as alternative to the original algorithms, for the fast computation of the Zernike and Pseudo-Zernike moments. Keywords: Orthogonal moments, Zernike moments, Pseudo-Zernike moments, direct, recursive algorithm. DOI: /S INTRODUCTION Image moments have been widely used as image descriptors in image processing and pattern classification applications. Chronologically, the firstly introduced moment type in image processing was the geometric moments [1]. Afterwards, since the geometric moments are not invariant to any image transformations (translation, rotation, scaling) the central and normalized moments [1 4] were introduced. However, since these types of moments are projections to a monomial, their bases are not orthogonal and thus there is a lot of information redundant and the reconstruction of the image by its moments is not possible. This deficiency of the conventional moments is not present in the orthogonal moments [1], which are projections of the image density function to orthogonal bases. The most significant property of the orthogonal moments is their ability to fully describe an object, with minimum redundant information and thus the reconstruction of an object by a finite number of moments, is possible. The orthogonal moments are categorized in families according to the type of the orthogonal polynomials that are making use, as kernel functions. The most utilized orthogonal families are the Zernike and 1 The article is published in the original. Received November 14, 007 Pseudo-Zernike moments [1], which are widely used as image descriptors in image processing tasks. Moreover, their property to remain invariant to any rotation of the object presented in a scene, in addition to their ability to describe spatial freuency components of an image, make them appropriate for pattern classification applications [5 8]. Although the previous orthogonal moments have useful attributes over the other moment types, the presence of many factorial terms in their polynomial definitions, make their computation a very time consuming task. While the computational capabilities of the modern computers are always increasing, the factorial of a big number remains a very demanding process. For this reason, many researchers have introduced recursive algorithms for the computation of Zernike [9 13] and Pseudo-Zernike [14] moments, by eliminating the factorial calculations. However, these algorithms have the possibility to generate and propagate finite precision errors, as classical signal processing algorithms do [15 17]. Moreover, the recursive algorithms that have been introduced until now, need the computation of intermediate polynomials of lower orders or repetitions. While this feature is an advantage for image representation and reconstruction [8 13] that accelerates their computation, in the case of pattern classification applications, where the computation of distinct moments of any order might be reuired, it decelerates the feature extraction process. ISSN , Pattern Recognition and Image Analysis, 010, Vol. 0, No. 1, pp Pleiades Publishing, Ltd., 010.

2 EFFICIENT COMPUTATION OF ZERNIKE AND PSEUDO-ZERNIKE MOMENTS 57 In the present paper, two novel algorithms that permit the computation of Zernike and Pseudo-Zernike moments of arbitrary orders and repetitions are introduced. These algorithms are recursive, do not generate and propagate finite precision errors and finally, allow single moments to be independently computed. The paper is organized as follows: in Section the fundamental theory about the orthogonal moments is presented. The discussion is focused on the Zernike and Pseudo-Zernike orthogonal families. In Section 3 the newly developed recursive algorithms are introduced while in Section 4 appropriate simulations establish the proposed algorithms as a general framework for computing the Zernike and Pseudo-Zernike moments. Some very useful conclusions are presented in Section 5.. ORTHOGONAL MOMENTS In this section a detailed description of the fundamental theory of the most utilized orthogonal families that are widely used in pattern recognition tasks, the Zernike and Pseudo-Zernike moments, is being given..1. Zernike Moments Zernike moments (ZMs) are the most widely used family of orthogonal moments due to their extra property of being invariant to an arbitrary rotation of the object they describe. They are used, after making them invariant to scale and translation, as object descriptors in pattern recognition applications [5 8] and in image retrieval tasks [18 0]. The introduction of ZMs in image analysis was made by Teague [1], using a set of complex polynomials, which form a complete orthogonal set over the interior of the unit circle x + y = 1. These polynomials [5, 6] have the form V p ( x, y) = V p ( r, θ) = R p () r exp( jθ), (1) where p is a non-negative integer and is a non zero integer subject to the constraints p even and p, r is the length of vector from the origin ( x, y ) to the pixel (x, y) and θ the angle between vector r and x axis in counter-clockwise direction. R nm (r), are the Zernike radial polynomials [] in (r, θ) polar coordinates defined as R p p () r = ( 1) s r ( s)! p s! p s! s = s! Note that R p, (r) = R p (r). s. () The polynomials of E. () are orthogonal and satisfy the orthogonality principle x + y 1 * ( x, y)v p ( x, y) dxdy = V nm π n δ + 1 δ, np m (3) where δ αβ = 1 for α = β and δ αβ = 0 otherwise, is the Kronecker symbol. The Zernike moment of order p with repetition for a continuous image function f(x, y), that vanishes outside the unit disk is n + 1 Z p = f( x, y)v p * ( ρθ, ) dxdy. π x + y 1 (4) For a digital image, the integrals are replaced by summations [5, 6] to get p + 1 Z p = f( x, y)v p * r, θ π ( ), x + y 1. x y (5) Suppose that one knows all moments Z p of f(x, y) up to a given order p max. It is desired to reconstruct a discrete function ˆ f (x, y) whose moments exactly match those of f(x, y) up to the given order p max. Zernike moments are the coefficients of the image expansion into orthogonal Zernike polynomials as can be seen in the following reconstruction euation, p max ˆ f ( x, y) = Zp V p ( r, θ), p = 0 (6) with having similar constraints as in (1). Note that as p max approaches infinity ˆ f (x, y) will approach f(x, y). As it can be seen from E. () there are a lot of factorial computations, operations that may consume too much computer time. For this reason, as it has already been discussed in the introduction, recursive algorithms for the computation of the radial polynomials () have been developed [9 13]. In the next section, a novel recursive algorithm for computing the radial polynomials () is presented. This algorithm doesn t include any factorial term that could cause any numerical instability and it computes the radial polynomials in a fast way... Pseudo-Zernike Moments Pseudo-Zernike moments (PZMs) are used in many pattern recognition [3, 4] and image processing applications as alternatives to the traditional ZMs. It has been proved that they have better feature representation capabilities and are more robust to image noise [5] than Zernike moments.

3 58 PAPAKOSTAS et al. The kernel of these moments is the orthogonal set of Pseudo-Zernike polynomials defined inside the unit circle. These polynomials have the form of (1) with the Zernike radial polynomials replaced by the Pseudo-Zernike radial polynomials () r S p = ( 1) s r ( p + 1 s)! p s! ( p+ + 1 s)! ( s)! s = 0 (7) with additional constraints 0 p, p = 0, 1,.,, (8) The corresponding PZMs are computed using the same formulas (4) and (5) as in the case of ZMs, since the only difference is in the form of the polynomial being used. Due to the above constraints (8), the set of Pseudo- Zernike polynomials of order p, contain (p + 1) linearly independent polynomials of degree p. On the other hand, the set of Zernike polynomials contain only (p + 1)(p + )/ linearly independent polynomials of degree p, due to the additional condition that p is even. Thus, PZMs offer more feature vectors than the Zernike moments of the same order. As can be seen from (7) the computation of Pseudo-Zernike moments, involves the calculation of some factorial terms, an operation that adds an extra computational effort. Conseuently, as in the case of Zernike moments, there is a demand for developing fast and numerical robust algorithms, able to calculate accurate moments in a short time. Similarly to ZMs, an image described by a finite number of PZMs, can be reconstructed by using E. (6). In the next section, recursive algorithms for the computation of ZMs and PZMs, which satisfy the previous reuirements, are presented and their computational complexity is being examined, in comparison to their original direct s. 3. THE PROPOSED ALGORITHMS From the previous sections, it is obvious that there is a need for developing algorithms that fast compute the Zernike and Pseudo-Zernike moments of arbitrary orders and repetitions, without generating and propagating numerical uantization errors. Such kind of algorithms would be very useful for pattern classification applications, where moments of any order and repetition may be used as discriminative features. Let us summarize some of the features that an efficient algorithm must have in order to be useful for pattern classification tasks: (1) Firstly, an algorithm for the computation of Zernike type moments, have to be numerically stable. This stability can be achieved by avoiding finite preci- s sion errors to be presented. This property is of major importance, since a possible numerical error generated in a step of the algorithm and propagated in subseuent steps, may cause the collapse of the algorithm. () Secondly, the algorithm has to be faster than its original direct version. This can be achieved by eliminating the need to compute factorial calculations, which adds a high computation effort. (3) Thirdly, the algorithm has to allow the computation of individual moments and radial polynomials, without intermediate calculations. The recursive algorithms introduced until now, compute the radial polynomials recursively. Thus, in order to compute a moment of a specific order and repetition, the calculation of radial polynomials of lower order or repetition is needed. This mechanism is desirable when the moments are used to reconstruct an image. However, for classification purposes, it is more useful to directly compute the radial polynomial of the order and repetition actually needed. This demand has been originated in the last years, where more sophisticated, evolutionary feature selection algorithms are applied, so that the most appropriate features for any application, is selected. A first attempt to use a genetic algorithm for the selection of the appropriate Zernike moments needed to be participant in a pattern recognition task as feature vectors, has been presented in [6]. Taking under consideration the above three reuirements, which should be satisfied by an algorithm for computing the ZMs and PZMs, in order to be useful for pattern classification applications, the following recursive algorithms are presented for both cases. The nature of the radial polynomials used in both cases, help us to manipulate them in the same way. The proof of the new algorithms is given in the Appendix. A. Zernike Moments We propose, the following formula (9) for the computation of the radial polynomial of order p and repetition, where the fractional terms are recursively computed using (10a) and (10b), where R p p, () r = T p0 r p + ( 1) k T pk r p k k = 1 T p! p0 = , ! p ! T k + 1 p k + 1 pk = T k( k+ 1) p k 1 ( ). (9) (10a) (10b)

4 EFFICIENT COMPUTATION OF ZERNIKE AND PSEUDO-ZERNIKE MOMENTS 59 Euation (10b), holds for k = 1,, 3,, (p )/, while for k = 0, one has to use (10a). B. Pseudo-Zernike Moments In the same way as ZMs, the radial polynomials used as kernel functions for the computation of PZMs, can take the form (a) (b) S p () r = Y p0 r p + ( 1) k Y pk r p k k = 1 (11) where Y p0 = , ( p + 1) (1a) ( p+ + 1)! ( )! ( p+ k+ ) ( k + 1) Y pk = Y (1b) k( k+ ) p ( k 1 ). Euation (1b), holds for k = 1,, 3, (p ), while for k = 0, one has to use (1a). The proposed algorithms are numerically robust in terms of finite precision errors, since the fundamental operations, subtraction, addition, multiplication and division are being made between integer and not real values. Therefore, according to [15 17], there is no possibility to produce finite precision errors, as a well known recursive algorithm for the computation of the Zernike moments [9] does, due to the subtraction operation [7]., 3.1. Computational Complexity The computational complexity of the proposed recursive algorithms, versus that of the direct s for the computation of the ZMs and PZMs, is being discussed in the current section. From Es. () and (7), owing to many factorial terms, it is obvious that the computational complexity of the direct s is very high Zernike moments. In the case of Zernike moments, the number of multiplications that have to be executed, in order to compute the radial polynomial using E. (), is almost (p/ + 1)*(p 3)*(p 1), in the worst case. The computational complexity of the direct, for computing a single radial polynomial R p of order p and repetition, is O(p 3 ). The proposed algorithm is of less computational complexity, since its recursive formula for the computation of the fractional terms, eliminates the factorials, by having only 6 multiplications in every fractional term, except the first one. The number of multiplications needed for the computation of a single radial polynomial using the proposed algorithm is approximately (3p 1) + p / + 4p, in the worst case. Thus, the computational complexity of the recursive algorithm, is O(p ). This complexity is smaller Fig. 1. Lena s grey level images of (a) and (b) sizes. than that of the original and makes the algorithm suitable for the computation of ZMs, up to very high orders. The efficiency of the algorithm in computing the ZMs, will be studied in section 4, where the moments of images of several sizes are being computed Pseudo-Zernike moments. Similarly to the Zernike moments, the Pseudo-Zernike moments have O(p ) computational complexity, as it can be observed by (7), where it is clear that the number of the reuired multiplications for computing the radial polynomial R p is approximately (p + 1)*(4p)*(p 1), where p is the moment order. The proposed recursive algorithm in the case of PZMs, has computational complexity O(p ), since the number of the multiplications needed to compute a radial polynomial of order p and repetition, R p is (5p + 1) + 6p + p, in the worst case. Conclusively, taking into account the previous observations about the computational complexity, we can claim that by applying a straightforward ology (see Appendix), very fast algorithms that permit the computation of individual moments of order p and repetition, can be achieved. In the following, the above detailed study of the computational complexity of the proposed recursive algorithms is justified, through appropriate experiments. 4. EXPERIMENTAL STUDY In order to study the effectiveness of the proposed recursive algorithms, a set of experiments has taken place. Two different sizes of the well known benchmark Lena s image have been selected, to be used as test images, with and pixels, in grey level format, as depicted in Fig. 1. The Zernike moments up to various maximum orders p max, are computed using E. () in the direct and E. (9) in the proposed. The CPU elapsed time for each one of the experiments has been measured and the results for both images' sizes are illustrated in Fig..

5 60 PAPAKOSTAS et al. ms Direct (a) Proposed Direct Proposed (b) Maximum moment otder Fig.. The CPU elapsed time (ms) for various maximum Zernike moment orders of (a) and (b) Lena s images. ms (a) Direct 4.0 Direct 10 Proposed 3.5 Proposed (b) Maximum moment otder Fig. 3. The CPU elapsed time (ms) for various maximum Pseudo-Zernike moment orders of (a) and (b) Lena s images. Figure 3 illustrates the CPU elapsed time needed to compute the Pseudo-Zernike moments of various maximum orders p max, for the two images when the direct and recursive algorithms are used. The above figures show that as the maximum moment order increases, the time needed to compute the moments, in the direct s is exponentially increased. The recursive algorithms, introduced in the current paper, need less CPU time, to compute the moments of the same order than the original ones. This observation, justifies experimentally, what has already been stated in Section 3, where the computational complexity of the algorithms has been proved to be of very low order. Apart from the above observations, Fig. and Fig. 3 show that our algorithms perform well, although the image size varies and thus their complexity is independent of the image size being processed. More precisely, the Zernike and Pseudo-Zernike moments for the two images and for maximum orders 10, 0, 30, 40, 50 have been computed and the CPU elapsed time in each case is presented in the following Table 1 and Table, respectively. The outperforming of the recursive s can also be defined if we calculate the % percentage of the time reduction taking place, as maximum order increases, by using the proposed s as alternatives of the direct ones. For this reason, we define the Computation Time Reduction (CTR) % as follows, Computation Time Reduction CRT % = Time Direct Time Direct Time Recursive (13)

6 EFFICIENT COMPUTATION OF ZERNIKE AND PSEUDO-ZERNIKE MOMENTS 61 Table 1. The CPU elapsed time (ms) for computation of Zernike moments up to p max order maximum order Lena Image direct recursive Lena Image direct recursive p max = p max = p max = p max = p max = Table. The CPU elapsed time (in seconds) for computation of Pseudo-Zernike moments up to p max order maximum order Lena Image direct recursive Lena Image direct recursive p max = p max = p max = p max = p max = By computing the computation time reduction using (13), for the same experiments of Fig. and Fig. 3, the following Fig. 4 can be drawn, Figure 4, verifies the benefits of the newly introduced algorithms, in terms of the computation time reduction, when being used instead of the direct s, for computing the Zernike and Pseudo-Zernike moments of any image size. The time reduction curves depicted in Fig. 4, for the two image sizes and for small moment orders, are not identical, as it was expected, due to the accuracy of the timer used to measure the computational time. When the computational time is uite short the accuracy of the timer significantly influences it and as the time increases this impact is negligible. From the above figure, it is impressive to conclude that we have a very significant computation time reduction, for both cases, which for high orders goes up to almost over 70%, in comparison with the original. This is a major advantage of the recursive algorithms, since in image representation one has to compute the moments of an image up to high orders, in order to optimally reconstruct it, with minimum reconstruction error. 5. CONCLUSIONS Two novel recursive algorithms were proposed in this paper, which compute in a fast way the orthogonal Zernike and Pseudo-Zernike moments. The computational complexity of the proposed algorithms is O(p ) in multiplications, while the original direct s are of O(p 3 ) complexity. Therefore, the computation time reuired to compute the moments of a high order, can be reduced by almost over 70% of the time needed by the direct s to do the same work. Additionally, the algorithms do not generate and propagate finite precision errors as some traditional recursive algorithms do. Also, they are capable to compute an individual moments for pattern classification purposes without needing the computation of intermediate moments. The experimental results justify the effectiveness of these new algorithms and establish them, as alternatives of the direct s, used until now. Percent of time reduction-% Lena image Lena image Maximum moment order (a) (b) Lena image Lena image Fig. 4. The CPU elapsed time (ms) for various maximum (a) Zernike and (b) Pseudo-Zernike moment orders of and Lena s images.

7 6 PAPAKOSTAS et al. APPENDIX DERIVATION OF THE RECURSIVE FORMULAS Let us define the fractional terms of order p, repetition and index k, which are used in the summations for computing the radial polynomials R p and S p of () and (7) respectively as follows, ( k)! T pk = (A.1) k! p k! p k! in the case of Zernike moments, and ( p + 1 k)! Y pk = (A.) k! ( p+ + 1 k)! ( k)! in the case of Pseudo-Zernike moments. The above euations (A.1) and (A.), may be combined to a more generic form as, ( a G 1 k)! pk = , (A.3) k! ( a k)! ( a 3 k)! where Zernike a 1 = p a = p a 3 = p Pseudo-Zernike a 1 = p + 1 a = p+ + 1 a 3 =. (A.4) The fractional term of (A.3) for the previous index of the summation (k 1), has the form, ( a 1 k + 1)! ( ) = ( k 1)! ( a k + 1)! ( a 3 k + 1)! G p k 1 (A.5) Now, if we restrict our study for indices k 0, we can use the property of the factorials n! = ( n 1)!n, n 0 (A.6) for transforming euation (A.5), as follows G p( k 1) = a 1 ( k)! ( a 1 k + 1)k k! ( a k)! ( a k + 1) ( a 3 k)! ( a 3 k + 1) ( a = 1 k)! k( a k + 1) (A.7) k! ( a k)! ( a k)! ( k + 1) ( a 3 k + 1) The above euation can be written, in a more suitable form, as ( a G k + 1) ( a 3 k + 1) pk = G (A.8) k( a 1 k + 1) p ( k 1 ). a k( a = T 1 k + 1) pk ( k + 1) ( a 3 k + 1) a By substituting the a 1, a, a 3 parameters in (A.8) using (A.4), we have in the case of Zernike Moments, p T k + 1 p k + 1 pk = T (A.9) k( k+ 1) p ( k 1 ), which holds for k = 1,, 3,, , and is identical to (10b) introduced in Section 3, and in the case of Pseudo-Zernike moments, Y pk = Y ( p+ k+ ) ( k + 1) (A.10) k( k+ ) p ( k 1 ), which holds for k = 1,, 3,, p, and is identical to (11b) introduced in Section 3. Finally, for k = 0, E. (A.3), gives a G 1! p0 = (A.11) a!a 3! In the same way, by substituting the a 1, a, a 3 parameters in (A.11) using (A.4), we have p! T p0 = (A.1) ! p ! and ( p + 1)! Y p0 = , (A.13) ( p+ + 1)! ( )! for the Zernike and Pseudo-Zernike moments, respectively. REFERENCES 1. R. Mukundan and K. R. Ramakrishnan, Moment Functions in Image Analysis (World Sci. Publ., 1998).. K. Tsirikolias and B. G. Mertizos, Statistical Pattern Recognition Using Efficient -D Moments with Applications to Character Recognition, Pattern Recognition 6, No. 6, (1993). 3. B. G. Mertzios and K. Tsirikolias, Statistical Shape Discrimination and Clustering Using an Efficient Set of Moments, Pattern Recognition Lett. 14, (June 1993). 4. B. G. Mertzios, Shape Discrimination in Robotic Vision Using Scaled Normalized Central Moments, in Proc. IFAC Workshop on Mutual Impact of Computing Power and Control Theory (Prague, Sept ), pp A. Khotanzad and J.-H. Lu, Classification of Invariant Image Representations Using a Neural Network, IEEE Trans. on Acoustics, Speech and Sign. Processing ASSP 38, No. 6, (1990). 6. G. A. Papakostas, Y. S. Boutalis, D. A. Karras, and B. G. Mertzios, A New Class of Zernike Moments for Computer Vision Applications, Information Sci. 177, No. 13, (007). 7. G. A. Papakostas, D. A. Karras, B. G. Mertzios, and Y. S. Boutalis, An Efficient Feature Extraction Methology for Computer Vision Applications Using Wavelet

8 EFFICIENT COMPUTATION OF ZERNIKE AND PSEUDO-ZERNIKE MOMENTS 63 Compressed Zernike Moments, ICGST Int. J. on Graphics, Vision and Image Processing, Special Issue: Wavelets and Their Applications SI1, 5 15 (005). 8. C. Kan and M. D. Srinath, Invariant Character Recognition with Zernike and Orthogonal Fourie Mellin Moments, Pattern Recognition 35 (1), (00). 9. C. W. Chong, P. Raveendran, and R. Mukundan, A Comparative Analysis of Algorithms for Fast Computation of Zernike Moments, Pattern Recognition 36 (3), (003). 10. S. O. Belkasim, M. Ahmadi, and M. Shridhar, Efficient Algorithm for Fast Computation of Zernike Moments, J. Franklin Inst. 333(B) (4), (1996). 11. J. Gu, H. Z. Shu, C. Toumoulin, and L. M. Luo, A Novel Algorithm for Fast Computation of Zernike Moments, Pattern Recognition 35 (1), (00). 1. A. Prata and W. V. T. Rusch, Algorithm for Computation of Zernike Polynomials Expansion Coefficients, Appl. Opt. 8, (1989). 13. E. C. Kintner, On the Mathematical Properties of the Zernike Polynomials, Opt. Acta 3 (8), (1976). 14. Chong Chee Way, R. Mucundan, and P. Raveendran, An Efficient Algorithm for Fast Computation of Pseudo Zernike Moments, in Proc. Int. Conf. on Image and Vision Computing (IVVCNZ 01) (Dunedin, 001), pp C. N. Papaodysseus, E. B. Koukoutsis, and C. N. Triantafyllou, Error Sources and Error Propagation in the Levinson Durbin Algorithm, IEEE Trans. Signal Processing SP 41 (4) (April 1993). 16. C. N. Papaodysseus, G. Carayannis, E. B. Koukoutsis, and E. Kayafas, Comparing LS FIR Filtering and 1-Step Ahead Linear Prediction, IEEE Trans. Signal Processing SP 41, No. (Feb. 1993). 17. C. Papaodysseus, E. Koukoutsis, and C. Vassilatos, Error Propagation and Methods of Error Correction in LS FIR, IEEE Trans. Signal Processing SP 4, No. 5 (May 1994). 18. T. W. Lin and Y. F. Chou, A Comparative Study of Zernike Moments for Image Retrieval, in Proc. 16th IPPR Conf. on Computer Vision, Graphics and Image Processing (CVGIP 003) (Taiwan, 003), pp M. Zhenjiang, Zernike Moment-Based Image Shape Analysis and its Application, Pattern Recognition Lett. 1 (), (000). 0. D. G. Sim, H. K. Kim, and R. H. Park, Invariant Texture Retrieval Using Modified Zernike Moments, Image and Vision Comp. (4), (004). 1. M. Teague, Image Analysis via the General Theory of Moments, J. Opt. Soc. Amer. 70 (8), (1980).. F. Zernike, Beugunstheorie des Schneidenverfahrens und seiner verbesserten Form, der Phasenkonstraste, Physica 1, (1934). 3. J. Haddadnia, M. Ahmadi, and K. Faez, An Efficient Feature Extraction Method with Pseudo-Zernike Moment in RBF Neural Network-Based Human Face Recognition System, EURASIP J. on Applied Signal Processing 9, (003). 4. J. Haddadnia, K. Faez, and M. Ahmadi, A Neural Based Human Face Recognition System Using an Efficient Feature Extraction Method with Pseudo Zernike Moment, J. Circuits, Systems, and Computers 11 (3), (00). 5. C.-H. The and R. T. Chin, On Image Analysis by the Methods of Moments, IEEE Trans. on Pattern Anal. Machine Intell. PAMI-10, No. 4, (1988). 6. G. A. Papakostas, Y. S. Boutalis, and B. G. Mertzios, Evolutionary Selection of Zernike Moment Sets in Image Processing, in Proc. 10th Intern. Workshop on Systems, Signals and Image Processing (IWSSIP 03) (Prague, Sept ). 7. G. A. Papakostas, Y. S. Boutalis, C. N. Papaodysseus, and D. K. Fragoulis, Numerical Error Analysis in Zernike Moments Computation, Image and Vision Comp. 4, (006). George A. Papakostas received the diploma of Electrical and Computer in Engineering in 1999 and the Msc. and PhD. Degree in Electrical and Computer Engineering (topic in Feature Extraction and Pattern Recognition) in 00 and 007 respectively, from Democritus University of Thrace (DUTH), Greece. He is the author of thirty publications in international scientific journals and conferences. He research interests are focused on the field of pattern recognition, neural networks, feature extraction, optimization, signal and image processing. Dr. Papakostas served as reviewer in numerous scientific journals and conferences and his is member of Technical Chamber of Greece. Yiannis S. Boutalis received the diploma of Electrical Engineer in 1983 from the Democritus University of Thrace (DUTH), Greece and the PhD degree in Electrical and Computer Engineering (topic Image Processing) in 1988 from the Computer Science Division of National Technical University of Athens, Greece. Since 1996, he serves as a faculty member, at the Department of Electrical and Computer Engineering, DUTH, Greece, where he is currently an Associate Professor and director of the automatic control systems lab. Currently, he is also a Visiting Professor for research cooperation at Erlangen-Nuremberg University of Germany, chair of Automatic Control. He served as an assistant visiting professor at University of Thessaly, Greece, and as a visiting professor in Air Defence Academy of General Staff of airforces of Greece. He also served as a researcher in the Institute of Language and Speech Processing (ILSP), Greece, and as a managing director of the R&D SME Ideatech S.A, Greece, specializing in pattern recognition and signal processing applications. His current research interests are focused in the development of Computational Intelligence techniues with applications in Control, Pattern Recognition, Signal and Image Processing Problems.

9 64 PAPAKOSTAS et al. Dimitrios A. Karras received his Diploma and M.Sc. Degree in Electrical and Electronic Engineering from the National Technical University of Athens, Greece in 1985 and the Ph. Degree in Electrical Engineering, from the National Technical University of Athens, Greece in 1995, with honors. In 1990 he received, also, his Diploma Degree in Mathematics from the University of Athens. Since 004, after his election, he has been with the Chalkis Institute of Technology, Automation Dept., Greece as full professor in Digital Systems as well as with the Hellenic Open University, Dept. Informatics as a visiting professor. He has published more than 40 research Journal papers in various areas of pattern recognition, image/signal processing and neural networks and more than 80 research papers in International Scientific Conferences. His research interests span the fields of pattern recognition and neural networks, multidimensional digital signal processing, image processing and analysis, communications and security, as well as parallel algorithms and fast processing. Basil G. Mertzios was born in Kavala, Greece on August 10, He received the Diploma Degree in Electrical and Mechanical Engineering from the Aristotle University of Thessaloniki, Greece in 1979 and the Ph. Degree in Electrical Engineering, from the Democritus University of Thrace, Greece in 198, both with honors. Part of his Ph. D. work has been done in University of Toronto. From he was an Associate Professor at the Democritus University of Thrace, Department of Electrical Engineering and Director of the Laboratory of Automatic Control Systems and from 199 a Professor at the same University. Also he has been invited and served as Professor of Control Systems and Computational Intelligence at the Department of Automation of the Technological Institute of Thessaloniki. During various periods since 1983, he visited extensively University of Toronto and Ruhr-Universitat Bochum, Germany as a research fellow. During part of 1986 he visited Georgia Institute of Technology, Atlanta, U.S.A., as an Adjunct Associate Professor. The academic year he was a Visiting Associate Professor at Georgia Institute of Technology, School of Electrical Engineering, in Fall Semester of 1991 he was a Visiting Professor at the University of Toronto and in 1996 he was a Visiting Professor at the Tokyo Institute of Technology, Japan. He was awarded in 1986 the Senior Fulbright Research Award, in 1987 the Alexander von Humboldt Fellowship and in 1988 the Natural Sciences and Engineering Research Council of Canada International Scientific Exchange Award. He published more than 100 research Journal papers in various areas of control system theory, multidimensional digital signal processing and robotics and more than 140 research papers in International Scientific Conferences. His research interests span the fields of automatic control systems, parallel algorithms and fast processing, robotics, multidimensional digital signal processing, image processing and analysis, pattern recognition and neural networks.

An Efficient Algorithm for Fast Computation of Pseudo-Zernike Moments

An Efficient Algorithm for Fast Computation of Pseudo-Zernike Moments An Efficient Algorithm for Fast Computation of Pseudo-Zernike Moments Chong, Chee-Way Fac. of Engg & Tech., Multimedia University, Jalan Air Keroh Lama, 75450 Melaka, Malaysia. (Email:cwayc@pc.jaring.my)

More information

MOMENT functions are used in several computer vision

MOMENT functions are used in several computer vision IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 13, NO. 8, AUGUST 2004 1055 Some Computational Aspects of Discrete Orthonormal Moments R. Mukundan, Senior Member, IEEE Abstract Discrete orthogonal moments

More information

Accurate Orthogonal Circular Moment Invariants of Gray-Level Images

Accurate Orthogonal Circular Moment Invariants of Gray-Level Images Journal of Computer Science 7 (5): 75-722, 20 ISSN 549-3636 20 Science Publications Accurate Orthogonal Circular Moment Invariants of Gray-Level Images Khalid Mohamed Hosny Department of Computer Science,

More information

FARSI CHARACTER RECOGNITION USING NEW HYBRID FEATURE EXTRACTION METHODS

FARSI CHARACTER RECOGNITION USING NEW HYBRID FEATURE EXTRACTION METHODS FARSI CHARACTER RECOGNITION USING NEW HYBRID FEATURE EXTRACTION METHODS Fataneh Alavipour 1 and Ali Broumandnia 2 1 Department of Electrical, Computer and Biomedical Engineering, Qazvin branch, Islamic

More information

Science Insights: An International Journal

Science Insights: An International Journal Available online at http://www.urpjournals.com Science Insights: An International Journal Universal Research Publications. All rights reserved ISSN 2277 3835 Original Article Object Recognition using Zernike

More information

Image Recognition Using Modified Zernike Moments

Image Recognition Using Modified Zernike Moments Sensors & Transducers 204 by IFSA Publishing S. L. http://www.sensorsportal.com Image Recognition Using Modified ernike Moments Min HUANG Yaqiong MA and Qiuping GONG School of Computer and Communication

More information

Object Recognition Using Local Characterisation and Zernike Moments

Object Recognition Using Local Characterisation and Zernike Moments Object Recognition Using Local Characterisation and Zernike Moments A. Choksuriwong, H. Laurent, C. Rosenberger, and C. Maaoui Laboratoire Vision et Robotique - UPRES EA 2078, ENSI de Bourges - Université

More information

A Novel Activity Detection Method

A Novel Activity Detection Method A Novel Activity Detection Method Gismy George P.G. Student, Department of ECE, Ilahia College of,muvattupuzha, Kerala, India ABSTRACT: This paper presents an approach for activity state recognition of

More information

Michal Kuneš

Michal Kuneš A (Zernike) moment-based nonlocal-means algorithm for image denoising Michal Kuneš xkunes@utia.cas.cz ZOI UTIA, ASCR, Friday seminar 13.03.015 Introduction - Uses nonlocal (NL) means filter - Introduce

More information

Blur Image Edge to Enhance Zernike Moments for Object Recognition

Blur Image Edge to Enhance Zernike Moments for Object Recognition Journal of Computer and Communications, 2016, 4, 79-91 http://www.scirp.org/journal/jcc ISSN Online: 2327-5227 ISSN Print: 2327-5219 Blur Image Edge to Enhance Zernike Moments for Object Recognition Chihying

More information

Empirical Analysis of Rotation Invariance in Moment Coefficients

Empirical Analysis of Rotation Invariance in Moment Coefficients Empirical Analysis of Rotation Invariance in Moment Coefficients Sukhjeet K. Ranade Department of Computer Science Punjabi University Patiala Supriya Anand Department of Computer Science Punjabi University

More information

Fast Computation of Orthogonal Polar Harmonic Transforms

Fast Computation of Orthogonal Polar Harmonic Transforms Fast Computation of Orthogonal Polar Harmonic Transforms Thai V. Hoang, Salvatore Tabbone To cite this version: Thai V. Hoang, Salvatore Tabbone. Fast Computation of Orthogonal Polar Harmonic Transforms.

More information

Palmprint Verification with Moments

Palmprint Verification with Moments Palmprint Verification with Moments Ying-Han Pang Andrew T.B.J David N.C.L Hiew Fu San Faculty of Information Science and Technology Multimedia University Jalan Ayer Keroh Lama 75450, Melaka, Malaysia.

More information

Hu and Zernike Moments for Sign Language Recognition

Hu and Zernike Moments for Sign Language Recognition Hu and Zernike Moments for Sign Language Recognition K. C. Otiniano-Rodríguez 1, G. Cámara-Chávez 1, and D. Menotti 1 1 Computing Department, Federal University of Ouro Preto, Ouro Preto, MG, Brazil Abstract

More information

Invariant Image Watermark Using Zernike Moments

Invariant Image Watermark Using Zernike Moments 766 IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, VOL. 13, NO. 8, AUGUST 2003 Invariant Image Watermark Using Zernike Moments Hyung Shin Kim and Heung-Kyu Lee, Member, IEEE Abstract This

More information

Invariant Pattern Recognition using Dual-tree Complex Wavelets and Fourier Features

Invariant Pattern Recognition using Dual-tree Complex Wavelets and Fourier Features Invariant Pattern Recognition using Dual-tree Complex Wavelets and Fourier Features G. Y. Chen and B. Kégl Department of Computer Science and Operations Research, University of Montreal, CP 6128 succ.

More information

Spatial Light Modulator for wavefront correction 1. INTRODUCTION

Spatial Light Modulator for wavefront correction 1. INTRODUCTION Spatial Light Modulator for wavefront correction A. Vyas 1, 2, M. B. Roopashree 1, Ravinder Kumar Banyal 1, B Raghavendra Prasad 1 1 Indian Institute of Astrophysics, Bangalore-560034, India 2 Indian Institute

More information

On the independence of rotation moment invariants

On the independence of rotation moment invariants Pattern Recognition 33 (2000) 1405}1410 On the independence of rotation moment invariants Jan Flusser* Institute of Information Theory and Automation Academy of Sciences of the Czech Republic Pod voda&

More information

A Recognition System for 3D Embossed Digits on Non-Smooth Metallic Surface

A Recognition System for 3D Embossed Digits on Non-Smooth Metallic Surface 2011 International Conference on elecommunication echnology and Applications Proc.of CSI vol.5 (2011) (2011) IACSI Press, Singapore A Recognition System for 3D Embossed Digits on Non-Smooth Metallic Surface

More information

Fast and Automatic Watermark Resynchronization based on Zernike. Moments

Fast and Automatic Watermark Resynchronization based on Zernike. Moments Fast and Automatic Watermark Resynchronization based on Zernike Moments Xiangui Kang a, Chunhui Liu a, Wenjun Zeng b, Jiwu Huang a, Congbai Liu a a Dept. of ECE., Sun Yat-Sen Univ. (Zhongshan Univ.), Guangzhou

More information

Constraint relationship for reflectional symmetry and rotational symmetry

Constraint relationship for reflectional symmetry and rotational symmetry Journal of Electronic Imaging 10(4), 1 0 (October 2001). Constraint relationship for reflectional symmetry and rotational symmetry Dinggang Shen Johns Hopkins University Department of Radiology Baltimore,

More information

Persian Handwritten Word Recognition Using Zernike and Fourier Mellin Moments

Persian Handwritten Word Recognition Using Zernike and Fourier Mellin Moments SETIT 009 5 th International Conference: Sciences of Electronic, Technologies of Information and Telecommunications March -6, 009 TUNISIA Persian Handwritten Word Recognition Using Zernike and Fourier

More information

Empirical Analysis of Invariance of Transform Coefficients under Rotation

Empirical Analysis of Invariance of Transform Coefficients under Rotation International Journal of Engineering Research and Development e-issn: 2278-67X, p-issn: 2278-8X, www.ijerd.com Volume, Issue 5 (May 25), PP.43-5 Empirical Analysis of Invariance of Transform Coefficients

More information

Feature Extraction Using Zernike Moments

Feature Extraction Using Zernike Moments Feature Extraction Using Zernike Moments P. Bhaskara Rao Department of Electronics and Communication Engineering ST.Peter's Engineeing college,hyderabad,andhra Pradesh,India D.Vara Prasad Department of

More information

Various Shape Descriptors in Image Processing A Review

Various Shape Descriptors in Image Processing A Review Various Shape Descriptors in Image Processing A Review Preetika D Silva 1, P. Bhuvaneswari 2 1 PG Student, Department of Electronics and Communication, Rajarajeswari College of Engineering, Bangalore,

More information

Lifting Parameterisation of the 9/7 Wavelet Filter Bank and its Application in Lossless Image Compression

Lifting Parameterisation of the 9/7 Wavelet Filter Bank and its Application in Lossless Image Compression Lifting Parameterisation of the 9/7 Wavelet Filter Bank and its Application in Lossless Image Compression TILO STRUTZ Deutsche Telekom AG, Hochschule für Telekommunikation Institute of Communications Engineering

More information

Comparative study of global invariant. descriptors for object recognition

Comparative study of global invariant. descriptors for object recognition Author manuscript, published in "Journal of Electronic Imaging (2008) 1-35" Comparative study of global invariant descriptors for object recognition A. Choksuriwong, B. Emile, C. Rosenberger, H. Laurent

More information

A comparative analysis of algorithms for fast computation ofzernike moments

A comparative analysis of algorithms for fast computation ofzernike moments Pattern Recognition 36 (003) 731 74 www.elsevier.com/locate/patcog A comparative analysis of algorithms for fast computation ofzernike moments Chee-Way Chong a;, P. Raveendran b, R. Mukundan c a Faculty

More information

Face recognition using Zernike moments and Radon transform

Face recognition using Zernike moments and Radon transform Face recognition using Zernike moments and Radon transform Dr. Ali Mohammed Sahan 1, Kadhim Zghair Oudah 2, Adel Majed 3 1 IT Dept., Technical College of Management, Middle Technical University, Baghdad,

More information

3-D Projective Moment Invariants

3-D Projective Moment Invariants Journal of Information & Computational Science 4: * (2007) 1 Available at http://www.joics.com 3-D Projective Moment Invariants Dong Xu a,b,c,d,, Hua Li a,b,c a Key Laboratory of Intelligent Information

More information

A Genetic Algorithm Approach to the Problem of Factorization of General Multidimensional Polynomials

A Genetic Algorithm Approach to the Problem of Factorization of General Multidimensional Polynomials 16 IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I: FUNDAMENTAL THEORY AND APPLICATIONS, VOL. 50, NO. 1, JANUARY 2003 A Genetic Algorithm Approach to the Problem of Factorization of General Multidimensional

More information

ROTATION, TRANSLATION AND SCALING INVARIANT WATERMARKING USING A GENERALIZED RADON TRANSFORMATION

ROTATION, TRANSLATION AND SCALING INVARIANT WATERMARKING USING A GENERALIZED RADON TRANSFORMATION ROTATION, TRANSLATION AND SCALING INVARIANT WATERMARKING USING A GENERALIZED RADON TRANSFORMATION D. SIMITOPOULOS, A. OIKONOMOPOULOS, AND M. G. STRINTZIS Aristotle University of Thessaloniki Electrical

More information

International Journal of Computer Engineering and Applications, Volume XII, Special Issue, August 18, ISSN

International Journal of Computer Engineering and Applications, Volume XII, Special Issue, August 18,   ISSN International Journal of Computer Engineering and Applications, Volume XII, Special Issue, August 18, www.ijcea.com ISSN 2321-3469 CONTENT-BASED IMAGE RETRIEVAL USING ZERNIKE MOMENTS AND SURF Priyanka

More information

Shape Classification Using Zernike Moments

Shape Classification Using Zernike Moments Shape Classification Using Zernike Moments Michael Vorobyov icamp at University of California Irvine August 5, 20 Abstract Zernike moments have mathematical properties, make them ideal image features to

More information

Open Access Measuring Method for Diameter of Bearings Based on the Edge Detection Using Zernike Moment

Open Access Measuring Method for Diameter of Bearings Based on the Edge Detection Using Zernike Moment Send Orders for Reprints to reprints@benthamscience.ae 114 The Open Automation and Control Systems Journal, 2015, 7, 114-119 Open Access Measuring Method for Diameter of Bearings Based on the Edge Detection

More information

Object Recognition Using a Neural Network and Invariant Zernike Features

Object Recognition Using a Neural Network and Invariant Zernike Features Object Recognition Using a Neural Network and Invariant Zernike Features Abstract : In this paper, a neural network (NN) based approach for translation, scale, and rotation invariant recognition of objects

More information

Stable IIR Notch Filter Design with Optimal Pole Placement

Stable IIR Notch Filter Design with Optimal Pole Placement IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL 49, NO 11, NOVEMBER 2001 2673 Stable IIR Notch Filter Design with Optimal Pole Placement Chien-Cheng Tseng, Senior Member, IEEE, and Soo-Chang Pei, Fellow, IEEE

More information

θ = φ + φ / φ 2 JURNAL GENERIC Norsharina Et Al. extended to other unequal scaling invariant by integrating with Aspect Moment Invariants [9].

θ = φ + φ / φ 2 JURNAL GENERIC Norsharina Et Al. extended to other unequal scaling invariant by integrating with Aspect Moment Invariants [9]. Vol.6 No. (Januari 0) JURNAL GENERIC An Integrated Formulation of Zernike for Mining Insect Images Norsharina Abu Bakar, Siti Mariyam Shamsuddin, Maslina Darus Abstract This paper presents mathematical

More information

This is a repository copy of Evaluation of output frequency responses of nonlinear systems under multiple inputs.

This is a repository copy of Evaluation of output frequency responses of nonlinear systems under multiple inputs. This is a repository copy of Evaluation of output frequency responses of nonlinear systems under multiple inputs White Rose Research Online URL for this paper: http://eprintswhiteroseacuk/797/ Article:

More information

A New Efficient Method for Producing Global Affine Invariants

A New Efficient Method for Producing Global Affine Invariants A New Efficient Method for Producing Global Affine Invariants Esa Rahtu, Mikko Salo 2, and Janne Heikkilä Machine Vision Group, Department of Electrical and Information Engineering, P.O. Box 45, 94 University

More information

SIMPLE GABOR FEATURE SPACE FOR INVARIANT OBJECT RECOGNITION

SIMPLE GABOR FEATURE SPACE FOR INVARIANT OBJECT RECOGNITION SIMPLE GABOR FEATURE SPACE FOR INVARIANT OBJECT RECOGNITION Ville Kyrki Joni-Kristian Kamarainen Laboratory of Information Processing, Department of Information Technology, Lappeenranta University of Technology,

More information

Logo Recognition System Using Angular Radial Transform Descriptors

Logo Recognition System Using Angular Radial Transform Descriptors Journal of Computer Science 7 (9): 1416-1422, 2011 ISSN 1549-3636 2011 Science Publications Logo Recognition System Using Angular Radial Transform Descriptors Omar Mohammed Wahdan, Khairuddin Omar and

More information

IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS PART C: APPLICATIONS AND REVIEWS, VOL. 30, NO. 1, FEBRUARY I = N

IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS PART C: APPLICATIONS AND REVIEWS, VOL. 30, NO. 1, FEBRUARY I = N IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS PART C: APPLICATIONS AND REVIEWS, VOL. 30, NO. 1, FEBRUARY 2000 125 Translation, Rotation, and Scale-Invariant Object Recognition L. A. Torres-Méndez,

More information

Acomplex-valued harmonic with a time-varying phase is a

Acomplex-valued harmonic with a time-varying phase is a IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. 46, NO. 9, SEPTEMBER 1998 2315 Instantaneous Frequency Estimation Using the Wigner Distribution with Varying and Data-Driven Window Length Vladimir Katkovnik,

More information

Invariant Feature Extraction from Fingerprint Biometric Using Pseudo Zernike Moments

Invariant Feature Extraction from Fingerprint Biometric Using Pseudo Zernike Moments Invariant Feature Extraction from Biometric Using Pseudo Zernike Moments C.Lakshmi Deepika 1, Dr. A Kandaswamy 2, C. Vimal 3, and B. Sathish 4 1,2,3,4 Department of Biomedical Engineering, PSG College

More information

Filter Design for Linear Time Delay Systems

Filter Design for Linear Time Delay Systems IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. 49, NO. 11, NOVEMBER 2001 2839 ANewH Filter Design for Linear Time Delay Systems E. Fridman Uri Shaked, Fellow, IEEE Abstract A new delay-dependent filtering

More information

Study of Zernike moments using analytical Zernike polynomials

Study of Zernike moments using analytical Zernike polynomials Available online at www.pelagiaresearchlibrary.com Advances in Applied Science Research, 2012, 3 (1):583-590 ISSN: 0976-8610 CODEN (USA): AASRFC Study of Zernike moments using analytical Zernike polynomials

More information

AN INVESTIGATION ON EFFICIENT FEATURE EXTRACTION APPROACHES FOR ARABIC LETTER RECOGNITION

AN INVESTIGATION ON EFFICIENT FEATURE EXTRACTION APPROACHES FOR ARABIC LETTER RECOGNITION AN INVESTIGATION ON EFFICIENT FEATURE EXTRACTION APPROACHES FOR ARABIC LETTER RECOGNITION H. Aboaisha 1, Zhijie Xu 1, I. El-Feghi 2 z.xu@hud.ac.uk Idrisel@ee.edu.ly 1 School of Computing and Engineering,

More information

SPEECH CLASSIFICATION USING ZERNIKE MOMENTS

SPEECH CLASSIFICATION USING ZERNIKE MOMENTS SPEECH CLASSIFICATION USING ZERNIKE MOMENTS Manisha Pacharne 1 and Vidyavati S Nayak 2 1 Department of Computer Engineering, DIAT (DU), Pune-411025 (India) 1 ce9manisha@diat.ac.in 2 Armament Research &

More information

Old painting digital color restoration

Old painting digital color restoration Old painting digital color restoration Michail Pappas Ioannis Pitas Dept. of Informatics, Aristotle University of Thessaloniki GR-54643 Thessaloniki, Greece Abstract Many old paintings suffer from the

More information

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION Vol. III Estimation and Compensation of Nonlinear Perturbations by Disturbance Observers - Peter C.

CONTROL SYSTEMS, ROBOTICS, AND AUTOMATION Vol. III Estimation and Compensation of Nonlinear Perturbations by Disturbance Observers - Peter C. ESTIMATION AND COMPENSATION OF NONLINEAR PERTURBATIONS BY DISTURBANCE OBSERVERS Peter C. Müller University of Wuppertal, Germany Keywords: Closed-loop control system, Compensation of nonlinearities, Disturbance

More information

Copy-Move Forgery Detection Using Zernike and Pseudo Zernike Moments

Copy-Move Forgery Detection Using Zernike and Pseudo Zernike Moments 039 The International Arab Journal of Information Technology, Vol. 13, No. 6A, 2016 Copy-Move Forgery Detection Using Zernike and Pseudo Zernike Moments Khaled Mahmoud and Arwa Abu-AlRukab Computer Science

More information

ARTIFICIAL NEURAL NETWORKS گروه مطالعاتي 17 بهار 92

ARTIFICIAL NEURAL NETWORKS گروه مطالعاتي 17 بهار 92 ARTIFICIAL NEURAL NETWORKS گروه مطالعاتي 17 بهار 92 BIOLOGICAL INSPIRATIONS Some numbers The human brain contains about 10 billion nerve cells (neurons) Each neuron is connected to the others through 10000

More information

On the Behavior of Information Theoretic Criteria for Model Order Selection

On the Behavior of Information Theoretic Criteria for Model Order Selection IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. 49, NO. 8, AUGUST 2001 1689 On the Behavior of Information Theoretic Criteria for Model Order Selection Athanasios P. Liavas, Member, IEEE, and Phillip A. Regalia,

More information

Determining the Optimal Decision Delay Parameter for a Linear Equalizer

Determining the Optimal Decision Delay Parameter for a Linear Equalizer International Journal of Automation and Computing 1 (2005) 20-24 Determining the Optimal Decision Delay Parameter for a Linear Equalizer Eng Siong Chng School of Computer Engineering, Nanyang Technological

More information

Original Article Design Approach for Content-Based Image Retrieval Using Gabor-Zernike Features

Original Article Design Approach for Content-Based Image Retrieval Using Gabor-Zernike Features International Archive of Applied Sciences and Technology Volume 3 [2] June 2012: 42-46 ISSN: 0976-4828 Society of Education, India Website: www.soeagra.com/iaast/iaast.htm Original Article Design Approach

More information

CHAPTER 1. Over 50 Years of Image Moments and Moment Invariants

CHAPTER 1. Over 50 Years of Image Moments and Moment Invariants CHAPTER 1 Over 50 Years of Image Moments and Moment Invariants George A. Papakostas This chapter aims to analyze the research eld of moments and moment invariants in a holistic way. Initially, a literature

More information

Multiscale Autoconvolution Histograms for Affine Invariant Pattern Recognition

Multiscale Autoconvolution Histograms for Affine Invariant Pattern Recognition Multiscale Autoconvolution Histograms for Affine Invariant Pattern Recognition Esa Rahtu Mikko Salo Janne Heikkilä Department of Electrical and Information Engineering P.O. Box 4500, 90014 University of

More information

A Robust Modular Wavelet Network Based Symbol Classifier

A Robust Modular Wavelet Network Based Symbol Classifier A Robust Modular Wavelet Network Based Symbol Classifier A.K. Mishra, P.W. Fieguth, D.A. Clausi University of Waterloo, Vision and Image Processing Group akmishra, pfieguth, dclausi@uwaterloo.ca Abstract.

More information

where =0,, 1, () is the sample at time index and is the imaginary number 1. Then, () is a vector of values at frequency index corresponding to the mag

where =0,, 1, () is the sample at time index and is the imaginary number 1. Then, () is a vector of values at frequency index corresponding to the mag Efficient Discrete Tchebichef on Spectrum Analysis of Speech Recognition Ferda Ernawan and Nur Azman Abu Abstract Speech recognition is still a growing field of importance. The growth in computing power

More information

The Efficient Discrete Tchebichef Transform for Spectrum Analysis of Speech Recognition

The Efficient Discrete Tchebichef Transform for Spectrum Analysis of Speech Recognition The Efficient Discrete Tchebichef Transform for Spectrum Analysis of Speech Recognition Ferda Ernawan 1, Nur Azman Abu 2 and Nanna Suryana 2 1 Faculty of Information and Communication Technology, Universitas

More information

Multiple Similarities Based Kernel Subspace Learning for Image Classification

Multiple Similarities Based Kernel Subspace Learning for Image Classification Multiple Similarities Based Kernel Subspace Learning for Image Classification Wang Yan, Qingshan Liu, Hanqing Lu, and Songde Ma National Laboratory of Pattern Recognition, Institute of Automation, Chinese

More information

Closed-Form Design of Maximally Flat IIR Half-Band Filters

Closed-Form Design of Maximally Flat IIR Half-Band Filters IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II: ANALOG AND DIGITAL SIGNAL PROCESSING, VOL. 49, NO. 6, JUNE 2002 409 Closed-Form Design of Maximally Flat IIR Half-B Filters Xi Zhang, Senior Member, IEEE,

More information

Utilizing the discrete orthogonality of Zernike functions in corneal measurements

Utilizing the discrete orthogonality of Zernike functions in corneal measurements Utilizing the discrete orthogonality of Zernike functions in corneal measurements Zoltán Fazekas 1, Alexandros Soumelidis 1, Ferenc Schipp 2 1 Systems and Control Lab, Computer and Automation Research

More information

Wavelet-based Salient Points with Scale Information for Classification

Wavelet-based Salient Points with Scale Information for Classification Wavelet-based Salient Points with Scale Information for Classification Alexandra Teynor and Hans Burkhardt Department of Computer Science, Albert-Ludwigs-Universität Freiburg, Germany {teynor, Hans.Burkhardt}@informatik.uni-freiburg.de

More information

A Modified Moment-Based Image Watermarking Method Robust to Cropping Attack

A Modified Moment-Based Image Watermarking Method Robust to Cropping Attack A Modified Moment-Based Watermarking Method Robust to Cropping Attack Tianrui Zong, Yong Xiang, and Suzan Elbadry School of Information Technology Deakin University, Burwood Campus Melbourne, Australia

More information

Image Interpolation Using Kriging Technique for Spatial Data

Image Interpolation Using Kriging Technique for Spatial Data Image Interpolation Using Kriging Technique for Spatial Data Firas Ajil Jassim, Fawzi Hasan Altaany Abstract Image interpolation has been used spaciously by customary interpolation techniques. Recently,

More information

Title without the persistently exciting c. works must be obtained from the IEE

Title without the persistently exciting c.   works must be obtained from the IEE Title Exact convergence analysis of adapt without the persistently exciting c Author(s) Sakai, H; Yang, JM; Oka, T Citation IEEE TRANSACTIONS ON SIGNAL 55(5): 2077-2083 PROCESS Issue Date 2007-05 URL http://hdl.handle.net/2433/50544

More information

A Recurrent Neural Network for Solving Sylvester Equation With Time-Varying Coefficients

A Recurrent Neural Network for Solving Sylvester Equation With Time-Varying Coefficients IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL 13, NO 5, SEPTEMBER 2002 1053 A Recurrent Neural Network for Solving Sylvester Equation With Time-Varying Coefficients Yunong Zhang, Danchi Jiang, Jun Wang, Senior

More information

Case Studies of Logical Computation on Stochastic Bit Streams

Case Studies of Logical Computation on Stochastic Bit Streams Case Studies of Logical Computation on Stochastic Bit Streams Peng Li 1, Weikang Qian 2, David J. Lilja 1, Kia Bazargan 1, and Marc D. Riedel 1 1 Electrical and Computer Engineering, University of Minnesota,

More information

A METHOD OF FINDING IMAGE SIMILAR PATCHES BASED ON GRADIENT-COVARIANCE SIMILARITY

A METHOD OF FINDING IMAGE SIMILAR PATCHES BASED ON GRADIENT-COVARIANCE SIMILARITY IJAMML 3:1 (015) 69-78 September 015 ISSN: 394-58 Available at http://scientificadvances.co.in DOI: http://dx.doi.org/10.1864/ijamml_710011547 A METHOD OF FINDING IMAGE SIMILAR PATCHES BASED ON GRADIENT-COVARIANCE

More information

Trademarks Cassification by Moment Invariant and FuzzyARTMAP

Trademarks Cassification by Moment Invariant and FuzzyARTMAP Trademarks Classification by Moment Invariant 1st International Workshop on Artificial Life and Robotics and FuzzyARTMAP 117-122 Trademarks Cassification by Moment Invariant and FuzzyARTMAP 1 Shahrul Nizam

More information

Boundary and Region based Moments Analysis for Image Pattern Recognition

Boundary and Region based Moments Analysis for Image Pattern Recognition ISSN 1746-7659, England, UK Journal of Inforation and Coputing Science Vol. 10, No. 1, 015, pp. 040-045 Boundary and Region based Moents Analysis for Iage Pattern Recognition Dr. U Ravi Babu 1, Dr Md Mastan

More information

Intermediate Algebra

Intermediate Algebra Intermediate Algebra 978-1-63545-084-2 To learn more about all our offerings Visit Knewton.com Source Author(s) (Text or Video) Title(s) Link (where applicable) Openstax Lyn Marecek, MaryAnne Anthony-Smith

More information

College Algebra with Corequisite Support: A Compressed Approach

College Algebra with Corequisite Support: A Compressed Approach College Algebra with Corequisite Support: A Compressed Approach 978-1-63545-059-0 To learn more about all our offerings Visit Knewton.com Source Author(s) (Text or Video) Title(s) Link (where applicable)

More information

DURING THE last two decades, many authors have

DURING THE last two decades, many authors have 614 IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I: FUNDAMENTAL THEORY AND APPLICATIONS, VOL 44, NO 7, JULY 1997 Stability the Lyapunov Equation for -Dimensional Digital Systems Chengshan Xiao, David J Hill,

More information

Algebra Performance Level Descriptors

Algebra Performance Level Descriptors Limited A student performing at the Limited Level demonstrates a minimal command of Ohio s Learning Standards for Algebra. A student at this level has an emerging ability to A student whose performance

More information

ALGEBRAIC CONDITION FOR DECOMPOSITION OF LARGE SCALE LINEAR DYNAMIC SYSTEMS

ALGEBRAIC CONDITION FOR DECOMPOSITION OF LARGE SCALE LINEAR DYNAMIC SYSTEMS Int. J. Appl. Math. Comput. Sci., 2009, Vol. 19, No. 1, 107 111 DOI: 10.2478/v10006-009-0010-x ALGEBRAIC CONDITION FOR DECOMPOSITION OF LARGE SCALE LINEAR DYNAMIC SYSTEMS HENRYK GÓRECKI Faculty of Informatics

More information

Role of Assembling Invariant Moments and SVM in Fingerprint Recognition

Role of Assembling Invariant Moments and SVM in Fingerprint Recognition 56 Role of Assembling Invariant Moments SVM in Fingerprint Recognition 1 Supriya Wable, 2 Chaitali Laulkar 1, 2 Department of Computer Engineering, University of Pune Sinhgad College of Engineering, Pune-411

More information

Elementary and Intermediate Algebra

Elementary and Intermediate Algebra Elementary and Intermediate Algebra 978-1-63545-106-1 To learn more about all our offerings Visit Knewton.com Source Author(s) (Text or Video) Title(s) Link (where applicable) OpenStax Lynn Marecek, Santa

More information

State Analysis and Optimal Control of Linear. Time-Invariant Scaled Systems Using. the Chebyshev Wavelets

State Analysis and Optimal Control of Linear. Time-Invariant Scaled Systems Using. the Chebyshev Wavelets Contemporary Engineering Sciences, Vol. 5, 2012, no. 2, 91-105 State Analysis and Optimal Control of Linear Time-Invariant Scaled Systems Using the Chebyshev Wavelets M. R. Fatehi a, M. Samavat b *, M.

More information

College Algebra with Corequisite Support: Targeted Review

College Algebra with Corequisite Support: Targeted Review College Algebra with Corequisite Support: Targeted Review 978-1-63545-056-9 To learn more about all our offerings Visit Knewtonalta.com Source Author(s) (Text or Video) Title(s) Link (where applicable)

More information

Rotational Invariants for Wide-baseline Stereo

Rotational Invariants for Wide-baseline Stereo Rotational Invariants for Wide-baseline Stereo Jiří Matas, Petr Bílek, Ondřej Chum Centre for Machine Perception Czech Technical University, Department of Cybernetics Karlovo namesti 13, Prague, Czech

More information

Watermarking in the Space/Spatial-Frequency Domain Using Two-Dimensional Radon Wigner Distribution

Watermarking in the Space/Spatial-Frequency Domain Using Two-Dimensional Radon Wigner Distribution 650 IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 10, NO. 4, APRIL 2001 Watermarking in the Space/Spatial-Frequency Domain Using Two-Dimensional Radon Wigner Distribution Srdjan Stanković, Member, IEEE,

More information

1946 IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 16, NO. 8, AUGUST 2007

1946 IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 16, NO. 8, AUGUST 2007 1946 IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 16, NO. 8, AUGUST 2007 Interconversion Between Truncated Cartesian and Polar Expansions of Images Wooram Park, Student Member, IEEE, and Gregory S. Chirikjian,

More information

Design of Orthonormal Wavelet Filter Banks Using the Remez Exchange Algorithm

Design of Orthonormal Wavelet Filter Banks Using the Remez Exchange Algorithm Electronics and Communications in Japan, Part 3, Vol. 81, No. 6, 1998 Translated from Denshi Joho Tsushin Gakkai Ronbunshi, Vol. J80-A, No. 9, September 1997, pp. 1396 1402 Design of Orthonormal Wavelet

More information

INSERM : LABORATOIRE INTERNATIONAL ASSOCI É, Universit é de Rennes I, SouthEast University, Rennes,FR

INSERM : LABORATOIRE INTERNATIONAL ASSOCI É, Universit é de Rennes I, SouthEast University, Rennes,FR Combined invariants to similarity transformation and to blur using orthogonal Zernike moments Chen Beijing 1 2, Huazhong Shu 1 2 *, Hui Zhang 1 2, Gouenou Coatrieux 3, Limin Luo 1 2 *, Jean-Louis Coatrieux

More information

Multisets mixture learning-based ellipse detection

Multisets mixture learning-based ellipse detection Pattern Recognition 39 (6) 731 735 Rapid and brief communication Multisets mixture learning-based ellipse detection Zhi-Yong Liu a,b, Hong Qiao a, Lei Xu b, www.elsevier.com/locate/patcog a Key Lab of

More information

The New Graphic Description of the Haar Wavelet Transform

The New Graphic Description of the Haar Wavelet Transform he New Graphic Description of the Haar Wavelet ransform Piotr Porwik and Agnieszka Lisowska Institute of Informatics, Silesian University, ul.b dzi ska 39, 4-00 Sosnowiec, Poland porwik@us.edu.pl Institute

More information

Using Entropy and 2-D Correlation Coefficient as Measuring Indices for Impulsive Noise Reduction Techniques

Using Entropy and 2-D Correlation Coefficient as Measuring Indices for Impulsive Noise Reduction Techniques Using Entropy and 2-D Correlation Coefficient as Measuring Indices for Impulsive Noise Reduction Techniques Zayed M. Ramadan Department of Electronics and Communications Engineering, Faculty of Engineering,

More information

A new method on deterministic construction of the measurement matrix in compressed sensing

A new method on deterministic construction of the measurement matrix in compressed sensing A new method on deterministic construction of the measurement matrix in compressed sensing Qun Mo 1 arxiv:1503.01250v1 [cs.it] 4 Mar 2015 Abstract Construction on the measurement matrix A is a central

More information

College Algebra with Corequisite Support: A Blended Approach

College Algebra with Corequisite Support: A Blended Approach College Algebra with Corequisite Support: A Blended Approach 978-1-63545-058-3 To learn more about all our offerings Visit Knewtonalta.com Source Author(s) (Text or Video) Title(s) Link (where applicable)

More information

A Method for Blur and Similarity Transform Invariant Object Recognition

A Method for Blur and Similarity Transform Invariant Object Recognition A Method for Blur and Similarity Transform Invariant Object Recognition Ville Ojansivu and Janne Heikkilä Machine Vision Group, Department of Electrical and Information Engineering, University of Oulu,

More information

IN many image processing applications involving wavelets

IN many image processing applications involving wavelets IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 17, NO. 7, JULY 2008 1061 Phase-Shifting for Nonseparable 2-D Haar Wavelets Mais Alnasser and Hassan Foroosh, Senior Member, IEEE Abstract In this paper, we

More information

Motion estimations based on invariant moments for frames interpolation in stereovision

Motion estimations based on invariant moments for frames interpolation in stereovision Available online at www.sciencedirect.com ScienceDirect Procedia Computer Science (013 ) 110 1111 17 th International Conference in Knowledge Based and Intelligent Information and Engineering Systems -

More information

Stability of Recursive Gaussian Filtering for Piecewise Linear Bilateral Filtering

Stability of Recursive Gaussian Filtering for Piecewise Linear Bilateral Filtering Stability of Recursive Gaussian Filtering for Piecewise Linear Bilateral Filtering Koichiro Watanabe, Yoshihiro Maeda, and Norishige Fukushima Nagoya Institute of Technology, Nagoya, Japan fukushima@nitech.ac.jp

More information

Invariant and Zernike Based Offline Handwritten Character Recognition

Invariant and Zernike Based Offline Handwritten Character Recognition International Journal of Advanced Research in Computer Engineering & Technology (IJARCET Volume 3 Issue 5, May 014 Invariant and Zernike Based Offline Handwritten Character Recognition S.Sangeetha Devi

More information

Analysis of Algorithm Efficiency. Dr. Yingwu Zhu

Analysis of Algorithm Efficiency. Dr. Yingwu Zhu Analysis of Algorithm Efficiency Dr. Yingwu Zhu Measure Algorithm Efficiency Time efficiency How fast the algorithm runs; amount of time required to accomplish the task Our focus! Space efficiency Amount

More information

The problem of classification in Content-Based Image Retrieval systems

The problem of classification in Content-Based Image Retrieval systems The problem of classification in Content-Based Image Retrieval systems Tatiana Jaworska Systems Research Institute, Polish Academy of Sciences Warsaw, Poland Presentation plan Overview of the main idea

More information

Algebra 1 Correlation of the ALEKS course Algebra 1 to the Washington Algebra 1 Standards

Algebra 1 Correlation of the ALEKS course Algebra 1 to the Washington Algebra 1 Standards Algebra 1 Correlation of the ALEKS course Algebra 1 to the Washington Algebra 1 Standards A1.1: Core Content: Solving Problems A1.1.A: Select and justify functions and equations to model and solve problems.

More information