• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.4
• /
• pp.343-348
• /
• 2014

Face recognition system has advanctage to automatically recognize a person without causing repulsion at deteciton process. However, the face recognition system has a drawback to show lower perfomance according to illumination variation unlike the other biometric systems using fingerprint and iris. Therefore, this paper proposed a robust face recogntion method against illumination varition by slective fusion technique using both visible and infrared faces based on fuzzy linear disciment analysis(fuzzy-LDA). In the first step, both the visible image and infrared image are divided into four bands using wavelet transform. In the second step, Euclidean distance is calculated at each subband. In the third step, recognition rate is determined at each subband using the Euclidean distance calculated in the second step. And then, weights are determined by considering the recognition rate of each band. Finally, a fusion face recognition is performed and robust recognition results are obtained.

• Journal of Korean Neurosurgical Society
• /
• v.38 no.5
• /
• pp.359-365
• /
• 2005

Objective : The study investigates the extent of brain shift and its effect on the accuracy of the stereotaxic procedure. Methods : Thirty-five patients underwent 40stereotactic procedures between June 2002 and March 2004. There were 26 males, mean age 59years old. There were 34procedures for Parkinson's disease, 2 for essential tremor, 3 for cerebral palsy, 1 for dystonia. Patients were divided in four groups based on postoperative pneumocephalus : under 5cc [9 procedures], between $5{\sim}10cc$ [13procedures], between $10{\sim}15cc$ [11 procedures] and more than 15cc [7procedures]. The coordinates of the anterior commissure[AC], posterior commissure[PC], and target were defined in pre-and intraoperative magnetic resonance image scans and the amount of air volume was measured with @Target (BrainLab, Heimstetten, Germany]. Results : The mean AC-PC was 26.5mm for patients with less than 5cc, 26.9mm for $5{\sim}10cc$, 25.8mm for $10{\sim}15cc$ and 26.2mm for more than 15cc. The length of AC-PC line and coordinates of AC, PC was also not statistically different, Euclidean distance as well as ${\Delta}x$, ${\Delta}y$, ${\Delta}z$ of AC, PC, and target were also not statistically different among the groups [p>,1]. There was a variance in target of $0.7{\sim}7.6mm$, Euclidean distance of 2.5mm, related to electrophysiology but not to brain-shift. Conclusion : The amount of air accumulated in the intracranial space and compressing the cortical surface has no effect on the localization of subcortical stereotactic target and landmarks.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.4
• /
• pp.715-720
• /
• 2016

In this paper, a method to design 3-dimensional (3-D) cross-lattice signal constellations with increased compactness is presented and analyzed. Here, the symbols located at the outermost sides in the conventional lattice constellation are moved symmetrically to fill in empty sides and sunken corners. While the minimum Euclidean distance (MED) among adjacent symbols remains unchanged, the presented cross-lattice constellations have 3~5% reduced average power and upto 25% reduced total volume as compared with the conventional ones. Due to the increase compactness, average power of the new 3-D constellations is lower than that of the conventional ones. As a result, computer simulation verifies that the presented cross-lattice constellations can improve symbol error performance of a digital transmission system about 0.4 [dB]. Hence, the proposed 3-D cross-lattice constellations are appropriate for low-power and high-quality digital communication systems.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.4
• /
• pp.427-431
• /
• 2010

Multidimensional scaling (MDS) is a widely used method for dimensionality reduction, of which purpose is to represent high-dimensional data in a low-dimensional space while preserving distances among objects as much as possible. MDS has mainly been applied to data visualization and feature selection. Among various MDS methods, the classical MDS is not readily applicable to data which has large numbers of objects, on normal desktop computers due to its computational complexity. More precisely, it needs to solve eigenpair problems on dissimilarity matrices based on Euclidean distance. Thus, running time and required memory of the classical MDS highly increase as n (the number of objects) grows up, restricting its use in large-scale domains. In this paper, we propose an efficient approximation algorithm for the classical MDS based on divide-and-conquer and CUDA. Through a set of experiments, we show that our approach is highly efficient and effective for analysis and visualization of data consisting of several thousands of objects.

• The Journal of the Acoustical Society of Korea
• /
• v.30 no.6
• /
• pp.303-307
• /
• 2011

This paper describes an audio watermarking using Quantization Index Modulation (QIM) on significant peaks in frequency domain. The audio signal is broken up into L samples length frames with non-overlapping and rectangular window. The zero-crossing rate of each frame is calculated for decision whether it is proper to be watermarked or not. If the frame is legitimate, frequency magnitude response is computed by discrete Fourier transform. For the QIM, we set the quantization step size based on maximum value of frequency magnitude response and select n significant peaks with w samples around them in frequency domain, totally $n{\times}(w+1)$ samples. Finally, watermark embedding is performed. Decoder extract watermarks based on Euclidean distance, that is a blind detection. The proposed method is robust against many attacks of watermark benchmark.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.5
• /
• pp.967-973
• /
• 2017

Clustering is one of the most well-known unsupervised learning methods that clusters data into homogeneous groups. Clustering has been used in various applications and FCM is one of the representative methods. In Fuzzy C-Means(FCM), however, cluster centers tend leaning to high density areas because the Euclidean distance measure forces high density clusters to make more contribution to clustering result. Previously proposed was density-independent clustering method, where cluster centers were made not to be close each other and relived the center deviation problem. Density-independent clustering method has a limitation that it is difficult to specify the position of the cluster centers. In this paper, an enhanced density-independent clustering method with an additional term that makes cluster centers to be placed around dense region is proposed. The proposed method converges more to real centers compared to FCM and density-independent clustering, which can be verified with experimental results.

• The KIPS Transactions:PartD
• /
• v.13D no.7 s.110
• /
• pp.879-888
• /
• 2006

Recently many researchers have focused on management of Historical trajectories of moving objects in Euclidean spaces due to numerous sizes of accumulated data over time. However, the movement of moving objects in real applications generally has some constraints, for example vehicles on roads can only travel along connected road networks. In this paper, we propose an indexing method for trajectories of moving objects on road networks in order to process the network-based spatiotemporal range query. Our method contains the connect information of road networks to use the network distance for query processing, deals with trajectories which are represented by road segments in road networks, and manages them using multiple R-trees assigned per each road segment. Furthermore, it has a structure to be able to share R-tree among several road segments in large road networks. Consequently, we show that our method takes about 30% less in node accesses for the network-based spatiotemporal range query processing than other methods based on the Euclidean distance by experiments.

• Journal of the Korean Society for Library and Information Science
• /
• v.41 no.2
• /
• pp.335-357
• /
• 2007

There has been many studies representing intellectual structures with multidimensional scaling(MDS) However MDS configuration is limited in representing local details and explicit structures. In this paper, we identified two components of MDS mapping approach; one is MDS algorithm and the other is preparation of data matrix. Various combinations of the two components of MDS mapping are compared through some measures of fit. It is revealed that the conventional approach composed of ALSCAL algorithm and Euclidean distance matrix calculated from Pearson's correlation matrix is the worst of the compared MDS mapping approaches. Otherwise the best approach to make MDS map is composed of PROXSCAL algorithm and z-scored Euclidean distance matrix calculated from Pearson's correlation matrix. These results suggest that we could obtain more detailed and explicit map of a knowledge domain through careful considerations on the process of MDS mapping.

• Journal of The Korean Society of Grassland and Forage Science
• /
• v.24 no.4
• /
• pp.293-300
• /
• 2004

This study was carried out to investigate the ecological behaviour forage value and similarity among the plant communities of the grazing pasture near Witzenhausen in middle part of Germany. Sixteen plant communities of the different grazing pasture were mostly the Molinio-Arrhenatheretea and Festuco-Brometea, and those were named the class of plant sociological nomenclature. The ecological behaviour and forage value of the communities except mesobromion(half dry grassland community) were relatively good for forage production. The correlation coefficient between class No. 14 and 12 of plant communities was highest, and the similarity among the communities were greatly affected by botanical composition. The resemblance measure of the cluster-analysis by complete-linkage-method for the similarity among plant communities was better the euclidean distance than those of others. The clustering analysis showed that the communities of relatively similar botanical composition were closely grouped.

• Journal of Korea Game Society
• /
• v.13 no.4
• /
• pp.15-24
• /
• 2013

The game of Go, which is originated at least more than 2,500 years ago, is one of the oldest board games in the world. So far the theoretical studies concerning to the Go openings are still insufficient. We applied traditional LDA algorithm to recognize a pro player's opening to a class obtained from the training openings. Both class-independent LDA and class-dependent LDA methods are conducted with the Go game records of the Korean top 10 professional Go players. Experimental result shows that the average recognition rate of class-independent LDA is 14% and class-dependent LDA 12%, respectively. Our research result also shows that in contrary to our common sense the algorithm based on PCA outperforms the algorithm based on LDA and reveals the new fact that the Euclidean distance metric method rarely does not inferior to LDA.