• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.5
• /
• pp.615-621
• /
• 2009

TFT-LCD image consists of ununiform background, random noises and target defect signal components. Defects in TFT-LCD have some intensity variations compared to background region. It is sometimes difficult for human inspectors to figure out. In this paper, we propose multi-level threshold scheme for detection of the real defect using probability density function with Parzen Window. The experimental results show that the proposed algorithms produce promising results and can be applied to automated inspection systems for finding defects in the TFT-LCD image.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.2
• /
• pp.196-201
• /
• 2007

The Fuzzy E-means (FCM) algorithm is a widely used clustering method that incorporates probabilitic memberships. Due to these memberships, it can be sensitive to noise data. In this paper, we propose a new fuzzy C-means clustering algorithm by incorporating the Parzen Window method to include density information of the data. Several experimental results show that our proposed density-based FCM algorithm outperforms conventional FCM especially for data with noise and it is not sensitive to initial cluster centers.

• Journal of Communications and Networks
• /
• v.12 no.5
• /
• pp.399-405
• /
• 2010

Blind equalization techniques have been used in broadcast and multipoint communications. In this paper, two criteria of minimizing Euclidian distance between two probability density functions (PDFs) for adaptive blind equalizers are presented. For PDF calculation, Parzen window estimator is used. One criterion is to use a set of randomly generated desired symbols at the receiver so that PDF of the generated symbols matches that of the transmitted symbols. The second method is to use a set of Dirac delta functions in place of the PDF of the transmitted symbols. From the simulation results, the proposed methods significantly outperform the constant modulus algorithm in multipath channel environments.

• Proceedings of the Korean Society of Soil and Groundwater Environment Conference
• /
• 2005.10a
• /
• pp.193-194
• /
• 2005

• Proceedings of the KSRS Conference
• /
• 2005.10a
• /
• pp.116-119
• /
• 2005

As to the synthetical estimation of land covering parameters or the compounded land covering classification for multi-resolution satellite data, former researches mainly adopted linear or nonlinear regression models to describe the regression relationship of land covering parameters caused by the degradation of spatial resolution, in order to improve the retrieval accuracy of global land covering parameters based on 1;he lower resolution satellite data. However, these methods can't authentically represent the complementary characteristics of spatial resolutions among different satellite data at arithmetic level. To resolve the problem above, a new compounded land covering classification method at arithmetic level for multi-resolution satellite data is proposed in this .paper. Firstly, on the basis of unsupervised clustering analysis of the higher resolution satellite data, the likelihood distribution scatterplot of each cover type is obtained according to multiple-to-single spatial correspondence between the higher and lower resolution satellite data in some local test regions, then Parzen window approach is adopted to derive the real likelihood functions from the scatterplots, and finally the likelihood functions are extended from the local test regions to the full covering area of the lower resolution satellite data and the global covering area of the lower resolution satellite is classified under the maximum likelihood rule. Some experimental results indicate that this proposed compounded method can improve the classification accuracy of large-scale lower resolution satellite data with the support of some local-area higher resolution satellite data.

• Proceedings of the IEEK Conference
• /
• 2002.07b
• /
• pp.810-813
• /
• 2002

We present the information theoretic learning based on the Tsallis entropy maximization principle for various q. The Tsallis entropy is one of the generalized entropies and is a canonical entropy in the sense of physics. Further, we consider the dependency of the learning on the parameter $\sigma$, which is a standard deviation of an assumed a priori distribution of samples such as Parzen window.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10a
• /
• pp.145-147
• /
• 2003

We derive a simple clustering algorithm which partitions the given data by minimizing overlap between clusters. For simple implementation and less complexity, Parzen window density estimation and quadratic distance measure between densities are adopted.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2006.11a
• /
• pp.211-214
• /
• 2006

Fuzzy C-Means(FCM) 알고리즘은 probabilitic 멤버쉽을 사용하는 클러스터링 방법으로서 널리 쓰이고 있다. 하지만 이 방법은 노이즈에 대하여 민감한 성질을 가진다는 단점이 있다. 따라서 본 논문에서는 이러한 노이즈에 민감한 성질을 보완하기 위해서 데이터의 밀도추정을 이용하여 새로운 FCM 알고리즘을 제안한다. 본 논문에서 제안된 알고리즘은 FCM과 비슷한 성능의 클러스터링 수행이 가능하며, 노이즈가 포함된 데이터에서는 FCM보다 더 나은 성능을 보여준다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.8 no.4
• /
• pp.75-88
• /
• 2008

This paper presents a spatial-level nonparametric multi-focus image fusion technique based on kernel estimates of input image blocks' underlying class-conditional probability density functions. Image fusion is approached as a classification task whose posterior class probabilities, P($wi{\mid}Bikl$), are calculated with likelihood density functions that are estimated from the training patterns. For each of the C input images Ii, the proposed method defines i classes wi and forms the fused image Z(k,l) from a decision map represented by a set of $P{\times}Q$ blocks Bikl whose features maximize the discriminant function based on the Bayesian decision principle. Performance of the proposed technique is evaluated in terms of RMSE and Mutual Information (MI) as the output quality measures. The width of the kernel functions, ${\sigma}$, were made to vary, and different kernels and block sizes were applied in performance evaluation. The proposed scheme is tested with C=2 and C=3 input images and results exhibited good performance.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.1C
• /
• pp.33-39
• /
• 2010

Correntropy is a generalized correlation function that contains higher order moments of the probability density function (PDF) than the conventional moment expansions. The criterion maximizing cross-correntropy (MCC) of two different random variables has yielded superior performance particularly in nonlinear, non-Gaussian signal processing comparing to mean squared error criterion. In this paper we propose a new blind equalization algorithm based on cross-correntropy criterion which uses, as two variables, equalizer output PDF and Parzen PDF estimate of a set of randomly generated symbols that complies with the transmitted symbol PDF. The performance of the proposed algorithm based on MCC is compared with the Euclidian distance minimization.