Y. He and A. Kundu, "2-d shape classification using hidden markov model," IEEE Transactions on Pattern Analysis Machine Intelligence, vol. 13, no. 11, pp. 1172--1184, Nov. 1991.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....Pavlidis and Ali [32] and Sze and Yang [37] Statistical techniques model shapes as one dimensional signals and attempt to estimate their parameters using autoregressive modeling and other statistical techniques. Examples include Das et al. 11] Dubois and Glanz [12] Eom [13] He and Kundu [18], and Kashyap and Chellappa [24] There are other shape analysis techniques (e.g. Gunsel and Telkap [17] and Neil and Curtis [29] that fall outside of these classes, and some that combine features of two or more classes (e.g. Cohen et al. 9] and Wang et al. 39] 4. COMPARISON METRICS ....

He, Y., and A. Kundu. 1991. "2-D Shape Classification Using Hidden Markov Model," IEEE Transactions on Pattern Analysis and Machine Intelligence PAMI13 (11):1172-1184 (November).

....2. Related works Bi dimensional shape recognition algorithms and methodologies are well documented in the literature. A lot of research has been carried out on this topic and has used many different approaches. Xu and Yang (1990) used multidimensional orthogonal polynomials, He and Kundu (1991), Sekita, Kurita and Otsu (1992) used autoregressive models. Neither of these approaches are well suited to grey scale images: in fact these methods take as input the boundary lines of objects. Other methods are based on the numerical Mellin transform of data. Zwicke and Kiss [8] used a Mellin ....

He Y. and Kundu A. (1991). "2-D Shape classification using hidden Markov model.", IEEE Tr.

....hand, many region based methods like [ 3,5,23,13,20,16,24,18,25,26,7,29,14,31,30 ] that use area surrounded by boundary curves concentrated just on finding reliable axis of shapes, which is another type of shapes rather than description usable by computers directly. Other region based methods [ 4,11,22,32,28,27 ] aimed at driving numerical representation by partitioning shapes into their parts based on the psychological observation [21] These methods require an analysis of its axis connections to find branch points where shapes are partitioned into their parts. Zhu [ 32 ] and Sharvit [27] used ....

Y. He and A. Kundu. "2-d shape classification using hidden markov model". IEEE Trans. Pattern. Anal. Mach. Intell., PAMI-13(11):1172--1184, 1991.

....While enormous amount of literature exists in each of the fields, these works have typically been developed in isolation. The following is a representative list: contour modeling [13, 33, 35, 48] edge detection [8, 11, 17, 30, 43] edge linking [3, 32, 34] contour detection and classification [18, 20, 36, 45]. This strict dichotomy of visual tasks has been widely adopted because it decomposes the problem of image analysis into independent and managable components. Unforturnately, the Marr s paradigm also results in unidirectional and irreversible flow of error from one component to the next. Moreover, ....

.... active contour model (snake) 21] parameterized templates [47] elliptic Fourier decompositions [41] implicit polynomials [42] handprinted character recognition [19] and affine invariant contour tracking [5] In addition, several methods that classify deformable contours have also been reported [18, 45], but these ignore the problem of extraction. There have been relatively few methods that consider detection and classification of deformable contours directly from noisy images [19, 42] 1.2 Contributions This thesis presents an integrated approach in modeling, extracting, detecting and ....

[Article contains additional citation context not shown here]

Y. He & A. Kundu, "2-D Shape Classification using Hidden Markov Model," IEEE Trans. Pat. Anal. Mach. Intell., vol. PAMI-13, 1991, pp. 1172-1184.

....shapes need to undergo much smaller metamorphosis to assume each other s shape than do dissimilar ones (see Fig. 1) Some of the other methods proposed by researchers for this problem include the use of joint probability density over shape to define model flexibility, Hidden Markov Model (HMM) [2, 3], multidimensional co occurrence matrices [4] elastic matching, and eigen mode based representation. Our approach differs from the research [2, 3, 4] in that it does not require extensive shape statistics. In real world applications, like user dependent pen based interfaces it allows the system ....

....proposed by researchers for this problem include the use of joint probability density over shape to define model flexibility, Hidden Markov Model (HMM) 2, 3] multidimensional co occurrence matrices [4] elastic matching, and eigen mode based representation. Our approach differs from the research [2, 3, 4] in that it does not require extensive shape statistics. In real world applications, like user dependent pen based interfaces it allows the system to perform with minimal training. Based on [7] we use a computationally inexpensive dynamic programming approach to find a globally optimal solution ....

[Article contains additional citation context not shown here]

Y. He and A. Kundu. "2D Shape Classification Using Hidden Markov Model". IEEE Transactions on Pattern Analysis and Machine Intelligence, 13(11):1172--1184, 1991.

....boundary extraction into Bayesian posterior estimation and thus exhibits high level of robustness and adaptability. This paper demonstrates the usefulness of g snake in classifying among several candidate deformable contours directly from noisy images. Although there already exist numerous methods [7, 8] that classify deformable contours, these typically assume that the contour has been previously extracted. However, without a contour model, contour extraction has been shown to be an illposed problem [9] Thus the practicality of these methods is limited in applications involving noisy images. ....

Y. He & A. Kundu, "2-D Shape Classification using Hidden Markov Model," IEEE Trans. Pat. Anal. Mach. Intell. , vol. PAMI-13, 1991, pp. 1172-1184.

....with occlusions and deformations, handling affine transformations, automatic processing, and optimal performance. Traditional approaches to shape comparison are quantitative in nature, using structural or stochastic comparison techniques to arrive at a measure of similarity or difference in shapes [2, 5, 3]. These methods provide little insight into the types and locations of differences between the shapes. Visualization techniques differ from quantitative techniques in that quantitative information is presented as pictures. Pictures convey qualitative information to the powerful pattern ....

He, Y., Kundu, A., "2-D Shape Classification Using Hidden Markov Model," IEEE Trans. Pattern Anal. Machine Intell., vol. 13, no. 11, Nov. 1991, pp. 1172-1184.

No context found.

Y. He and A. Kundu, "2-d shape classification using hidden markov model," IEEE Transactions on Pattern Analysis Machine Intelligence, vol. 13, no. 11, pp. 1172--1184, Nov. 1991.

No context found.

Y. He and A. Kundu, "2-d shape classification using hidden markov model," IEEE Trans. Pattern Analysis Machine Intelligence, vol. 13, no. 11, pp. 1172--1184, Noember. 1991.

No context found.

He, Y. & Kundu, A. "2D shape classification using hidden Markov model". In: IEEE Trans. Pattern Anal. Mach. Intell. PAMI-13. v.11, no. 13. Nov/1991. p. 1172-1184

No context found.

Y. He and A. Kundu, "2-D Shape Classification Using Hidden Markov Model," IEEE Trans. Pattern Anal. Mach. Intell. PAMI-13, 1172-1184 (1991).

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC