• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.38 no.6
• /
• pp.684-691
• /
• 2001

In this paper, we design and implement a system which recognizes and tracks a human head on a sequence of images. In this paper, the color of the skin and ellipse modeling is used as feature vectors to recognize the human head. And the modified time-varying edge detection method and the vertical projection method is used to acquire regions of the motion from images with very complex backgrounds. To select the head from the acquired candidate regions, the process for thresholding on the basis of the I-component of YIQ color information and mapping with ellipse modeling is used. The designed system shows an excellent performance in the cases of the rotated heads, occluded heads, and tilted heads as well as in the case of the normal up-right heads. And in this paper, the combinational technique of motion-based tracking and recognition-based tracking is used to track the human head exactly even though the human head moves fast.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.1
• /
• pp.24-31
• /
• 1994

In this paper, we report on the development of a speaker independent continuous speech recognition system using continuous hidden Markov models. The continuous hidden Markov model consists of mean and covariance matrices and directly models speech signal parameters, therefore does not have quantization error. Filter bank coefficients with their 1st and 2nd-order derivatives are used as feature vectors to represent the dynamic features of speech signal. We use the segmental K-means algorithm as a training algorithm and triphone as a recognition unit to alleviate performance degradation due to coarticulation problems critical in continuous speech recognition. Also, we use the one-pass search algorithm that Is advantageous in speeding-up the recognition time. Experimental results show that the system attains the recognition accuracy of $83\%$ without grammar and $94\%$ with finite state networks in speaker-indepdent speech recognition.

• The Journal of the Acoustical Society of Korea
• /
• v.11 no.3
• /
• pp.5-13
• /
• 1992

Taking devantage of the property that Korean digit is a mono-syllable word, we proposed a spoken Korean-digit recognition scheme using the multi-layer perceptron. The spoken Korean-digit is divided into three segments (initial sound, medial vowel, and final consonant) based on the voice starting / ending points and a peak point in the middle of vowel sound. The feature vectors such as cepstrum, reflection coefficients, ${\Delta}$cepstrum and ${\Delta}$energy are extracted from each segment. It has been shown that cepstrum, as an input vector to the neural network, gives higher recognition rate than reflection coefficients. Regression coefficients of cepstrum did not affect as much as we expected on the recognition rate. That is because, it is believed, we extracted features from the selected stationary segments of the input speech signal. With 150 ceptral coefficients obtained from each spoken digit, we achieved correct recognition rate of 97.8%.

• International Journal of Oral Biology
• /
• v.43 no.2
• /
• pp.101-111
• /
• 2018

Understanding the classification of malocclusion is a crucial issue in Orthodontics. It can also help us to diagnose, treat, and understand malocclusion to establish a standard for definite class of patients. Principal component analysis (PCA) and k-means algorithms have been emerging as data analytic methods for cephalometric measurements, due to their intuitive concepts and application potentials. This study analyzed the macro- and meso-scale classification structure and feature basis vectors of 1020 (415 male, 605 female; mean age, 25 years) orthodontic patients using statistical preprocessing, PCA, random matrix theory (RMT) and k-means algorithms. RMT results show that 7 principal components (PCs) are significant standard in the extraction of features. Using k-means algorithms, 3 and 6 clusters were identified and the axes of PC1~3 were determined to be significant for patient classification. Macro-scale classification denotes skeletal Class I, II, III and PC1 means anteroposterior discrepancy of the maxilla and mandible and mandibular position. PC2 and PC3 means vertical pattern and maxillary position respectively; they played significant roles in the meso-scale classification. In conclusion, the typical patient profile (TPP) of each class showed that the data-based classification corresponds with the clinical classification of orthodontic patients. This data-based study can provide insight into the development of new diagnostic classifications.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.4
• /
• pp.1033-1043
• /
• 2000

Recently as the user request on NoD (News-on-Demand) is largely increasing, there are a lot of researches to fulfill it. However, because of short life-cycle of new video data and periodical change of video data depending on anchor, it is difficult to apply the conventional video storage techniques to NOD applications directly. For this, we design and implement low-level storage manager for efficient storage and retrieval of multimedia data in NOD Services. Our low-level storage manager not only efficiently sotres video stream dat of new video itself, but also handles its index information. It provides an inverted file method for efficient text-based retrieval and an X-tree index structure for high-dimensional feature vectors. In addition, our low-level storage manager provides some application program interfaces (APIs) for storing video objects itself and index information extracted from hierarchial new video and some APIs for retrieving video objects easily by using cursors. Finally, we implement our low-level storage manager based on SHORE (Scalable Heterogeneous Object REpository) storage system by sunig a standard C++ language under UNIX operating system.

• The Journal of Korea Robotics Society
• /
• v.6 no.2
• /
• pp.130-140
• /
• 2011

This paper proposes an optimal posture for the task-oriented movement of dual arm manipulator. A stability criterion function which consists of three kinds of feature-representative parameters has been utilized to define the optimal posture. The first parameter is the force which is applied to the object. The torque of each joint and position of arm are attained from the current sensor and encoder, respectively. From these two data, the applied force to an object is estimated using sum of vectors of the joint torques estimated from the measured current. In order to investigate the robustness of each posture, the variation of the end-effector from the encoder information has been utilized as the second parameter. And for the last parameter for the optimality, the total energy consumption has been used. The total consuming energy of each posture can be computed from the current information and the battery voltage. The proposed robot structure consists of a mobile inverted pendulum and dual manipulators. In order to define the optimal posture for the each object, external disturbances are applied to the mobile inverted pendulum robot and the first and second parameters are investigated to find the optimal posture among the pre-selected most representative postures. Finally, the proposed optimal posture has been verified by the proposed stability criterion function which consists of total force to the object, the fluctuation of the end-effector position, and total energy consumption. The effectiveness of the proposed algorithms has been verified and demonstrated through the practical simulations and real experiments.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.25 no.6_1
• /
• pp.545-555
• /
• 2007

3D buildings are the most prominent feature comprising urban scene. A few of mega-cities in the globe are virtually reconstructed in photo-realistic 3D models, which becomes accessible by the public through the state-of-the-art online mapping services. A lot of research efforts have been made to develop automatic reconstruction technique of large-scale 3D building models from remotely sensed data. However, existing methods still produce irregular building polygons due to errors induced partly by uncalibrated sensor system, scene complexity and partly inappropriate sensor resolution to observed object scales. Thus, a geometric regularization technique is urgently required to rectify such irregular building polygons that are quickly captured from low sensory data. This paper aims to develop a new method for regularizing noise building outlines extracted from airborne LiDAR data, and to evaluate its performance in comparison with existing methods. These include Douglas-Peucker's polyline simplication, total least-squared adjustment, model hypothesis-verification, and rule-based rectification. Based on Minimum Description Length (MDL) principal, a new objective function, Geometric Minimum Description Length (GMDL), to regularize geometric noises is introduced to enhance the repetition of identical line directionality, regular angle transition and to minimize the number of vertices used. After generating hypothetical regularized models, a global optimum of the geometric regularity is achieved by verifying the entire solution space. A comparative evaluation of the proposed geometric regulator is conducted using both simulated and real building vectors with various levels of noise. The results show that the GMDL outperforms the selected existing algorithms at the most of noise levels.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.4
• /
• pp.1824-1845
• /
• 2016

The GMM is a conventional approach which has been recently applied in many face recognition studies. However, the question about how to deal with illumination changes while ensuring high performance is still a challenge, especially with real-world databases. In this paper, we propose a Visual Observation Confidence (VOC) measure for robust face recognition for illumination changes. Our VOC value is a combined confidence value of three measurements: Flatness Measure (FM), Centrality Measure (CM), and Illumination Normality Measure (IM). While FM measures the discrimination ability of one face, IM represents the degree of illumination impact on that face. In addition, we introduce CM as a centrality measure to help FM to reduce some of the errors from unnecessary areas such as the hair, neck or background. The VOC then accompanies the feature vectors in the EM process to estimate the optimal models by modified-GMM training. In the experiments, we introduce a real-world database, called KoFace, besides applying some public databases such as the Yale and the ORL database. The KoFace database is composed of 106 face subjects under diverse illumination effects including shadows and highlights. The results show that our proposed approach gives a higher Face Recognition Rate (FRR) than the GMM baseline for indoor and outdoor datasets in the real-world KoFace database (94% and 85%, respectively) and in ORL, Yale databases (97% and 100% respectively).

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.2
• /
• pp.149-157
• /
• 2011

Undecided inference could be regarded as a missing data problem such as MARand MNAR. Under the assumption of MAR, undecided inference make use of logistic regression model. The probability of default for the undecided group is obtained with regression coefficient vectors for the decided group and compare with the probability of default for the decided group. And under the assumption of MNAR, undecide dinference make use of logistic regression model with additional feature random vector. Simulation results based on two kinds of real data are obtained and compared. It is found that the misclassification rates are not much different from the rate of rawdata under the assumption of MAR. However the misclassification rates under the assumption of MNAR are less than those under the assumption of MAR, and as the ratio of the undecided group is increasing, the misclassification rates is decreasing.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.15 no.6
• /
• pp.579-589
• /
• 2004

The radar signals are generally very sensitive to relative orientations between radar and target. Thus, the performance of a target recognition system significantly deteriorates as the region of aspect angles becomes broader. To address this difficulty, in this paper, we propose a method based on the multi-aspect information in order to improve the classification capability ever for a wide angular region. First, range profiles are used to extract feature vectors based on the central moments and principal component analysis(PCA). Then, a classifier with the use of multi-aspect information is applied to them, yielding an additional improvement of target recognition capability. There are two different strategies among the classifiers that can fuse the information from multi-aspect radar signals: independent methodology and dependent methodology. In this study, the performances of the two strategies are compared within the frame work of target recognition. The radar cross section(RCS) data of six aircraft models measured at compact range of Pohang University of Science and Technology are used to demonstrate and compare the performances of the two strategies.