• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.342-347
• /
• 2008

Face tracking and recognition are difficult problems because the face is a non-rigid object. If the system tries to track or recognize the unknown face continuously, it can be more hard problems. In this paper, we propose the method to track and to recognize the face of the unknown person on video sequences using linear combination of nonlinear manifold models that is constructed in the system. The arbitrary input face has different similarities with different persons in system according to its shape or pose. Do we can approximate the new nonlinear manifold model for the input face by estimating the similarities with other faces statistically. The approximated model is updated at each frame for the input face. Our experimental results show that the proposed method is efficient to track and recognize for the arbitrary person.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.38 no.5
• /
• pp.23-32
• /
• 2001

Electrical impedance tomography (EIT) is a relatively new imaging modality in which the resistivity (conductivity) distribution of the unknown object is estimated based on the known sets of injected currents and measured voltages on the surface of the object. In this paper, we propose a dynamical EIT reconstruction algorithm based on the regularized extended Kalman filter(EKF). The EIT inverse problem is formulated as dynamic equation which consists of the slate equation and the observation equation, and the unknown state(resistivity) is estimated recursively with the aid of the EKF. In doing so, the generalized Tikhonov regularization technique is employed in the cost functional to mitigate the ill-posedness characteristics of the inverse problem. Computer simulations for the 16-channel synthetic data are provided to illustrate the reconstruction performance of the proposed algorithm.

• Journal of the Korean Society of Costume
• /
• v.6
• /
• pp.145-160
• /
• 1982

Wearing a clothes in human living is very rightful and reasonable things. And it's also the necessaries of living. So now, I want look around about origin of clothes from the prehistorical age, how they start and why they need to wear a clothes. As human culture is unknown origin, human clothes origin is also unknown origin, too. So many different people made unlike story about that, and it is very difficult to say correct birth of clothes. It must be reflection by a time, by a place, by a mankind, until this days, the history of human progress is $2,000,000{\sim}3,000,000$ years. And from after the "Old Adam" man understand what is shame and disgrace, so that made them cover of their private parts, also it is a motive of human clothes. Since after pass a long time, the earth has many changes as it were weather, and for take care of human body from cold, people start make clothes and it's a second object of wearing a clothes. And third object of wearing clothes is after opportunity of development human culture, people desire for an aesthetic sence and for full-fill that feeling they start make a beautiful clothes with accessories and it is motive and a purpose of a human start wear a clothes. So I can say, for origin birth of clothes is 1. Protect of human body. 1) be adapted for different weather and different circumstance. 2) take care self from outside injurg. 2. Decorate for aesthetic sence. 1) Theory of Amulets. 2) Theory of symbolism. 3) Theory of Sex attraction. 4) Aesthetic Theory. 5) Theory of chastity. 6) Multi-theory And material for clothes they made from natural wilds like leaves, fur, skin of wood. skin of fishes and feather etc. But after develop of human knowledge people invent fiber, and use a knitted goods and woven stuff also first of all. They start use those kind that they just cover of their private parts as a belt and astring with leaves and a beast skins. But until this days it's taken big development with human knowledge and most important influence of develop was weather. So modern clothes for present days, must care with. physicology, preservation of health, living activity, dressing, morals, and etiquette etc.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.11
• /
• pp.1448-1454
• /
• 2018

In an optical observing system, three pairs of observations at equal time intervals are required for the orbit determination method to obtain orbital information of an unknown space objects. In this paper, we propose a method of estimating a streak for acquiring three pairs of observations using one streak image information. Satellite trajectory simulation data were generated for nine cases using the STK program in order to verify the characteristics of the orbit of space object and estimation performance. Simulation was performed by applying three approaches that can estimate the next streak position after a few seconds from one streak image information, and the estimation performance was evaluated. Linear vector method and Kalman Filter method based on the linear assumption tend to increase the estimation error in the region where the nonlinearity is large. However estimation method using the polynomial curve fitting based on the least square method showed smaller and uniform error result than the previous methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.4
• /
• pp.938-958
• /
• 2024

Unsupervised Video Object Segmentation (UVOS) is a highly challenging problem in computer vision as the annotation of the target object in the testing video is unknown at all. The main difficulty is to effectively handle the complicated and changeable motion state of the target object and the confusion of similar background objects in video sequence. In this paper, we propose a novel deep Dual-stream Co-enhanced Network (DC-Net) for UVOS via bidirectional motion cues refinement and multi-level feature aggregation, which can fully take advantage of motion cues and effectively integrate different level features to produce high-quality segmentation mask. DC-Net is a dual-stream architecture where the two streams are co-enhanced by each other. One is a motion stream with a Motion-cues Refine Module (MRM), which learns from bidirectional optical flow images and produces fine-grained and complete distinctive motion saliency map, and the other is an appearance stream with a Multi-level Feature Aggregation Module (MFAM) and a Context Attention Module (CAM) which are designed to integrate the different level features effectively. Specifically, the motion saliency map obtained by the motion stream is fused with each stage of the decoder in the appearance stream to improve the segmentation, and in turn the segmentation loss in the appearance stream feeds back into the motion stream to enhance the motion refinement. Experimental results on three datasets (Davis2016, VideoSD, SegTrack-v2) demonstrate that DC-Net has achieved comparable results with some state-of-the-art methods.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.553-558
• /
• 2003

The boundary estimation problem is used to estimate the shape of organic depend on the phase of the cardiac cycle or interested in the detection of the location and size of anomalies with resistivity values different from the background tissues such as nuclear reactor. And we can use the method to solve the optimal solution such as modified Newton raphson, kalman filter, extended kalman filter, etc. But, this method consumes much time and is sensitive to the initial value and noise in the estimation of the unknown shape. In the paper, we propose that multi-layer neural networks estimate the boundary of the unknown object using Fourier coefficient. This method can be used at the real time estimation and have strong characteristics at the noise and initial value. It uses voltage change; difference the homogeneous voltage to the non-homogeneous voltage, and change of Fourier coefficient change to train multi-layer neural network. After train, we can have real time estimation using this method.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.570-574
• /
• 2003

In most boundary estimation algorithms estimation in EIT (Electrical Impedance Tomography), anomaly boundaries can be expressed with Fourier series and the unknown coefficients are estimated with proper inverse algorithms. Furthermore, the number of anomalies is assumed to be available a priori. The prior knowledge on the number of anomalies may be unavailable in some cases, and we need to determine the number of anomalies with other methods. This paper presents an algorithm for the boundary estimation in EIT (Electrical Impedance Tomography) using the prior information from the conventional Newton-Raphson method. Although Newton-Raphson method generates so poor spatial resolution that the anomaly boundaries are hardly reconstructed, even after a few iterations it can give general feature of the object to be imaged such as the number of anomalies, their sizes and locations, as long as the anomalies are big enough. Some numerical experiments indicate that the Newton-Raphson method can be used as a good predictor of the unknown boundaries and the proposed boundary discrimination algorithm has a good performance.

• The Bulletin of The Korean Astronomical Society
• /
• v.41 no.2
• /
• pp.48.3-49
• /
• 2016

We have five different types of observation modes with regard to the Deep Ecliptic Patrol of the Southern Sky (DEEP-South); Opposition Census (OC) for targeted photometry, Sweet Spot Survey (S1) for discovery and orbit characterization of Atens and Atiras, Ecliptic Survey (S2) for asteroid family studies and comet census, NEOWISE follow-up (NW) for near simultaneous albedo measurements in the visible bands, and Target of Opportunity (TO) observation for follow-up either for unpredictable events or targets of special interests. Different exposures with such different modes result in a wide range of background noise level, the number of background stars and the mover's projected speed in each image. The Moving Object Detection Program (MODP) utilizes multiple mosaic images being taken for the same target fields at different epochs at the three KMTNet sites. MODP employs existing software packages such as SExtractor (Source-Extractor) and SCAMP (Software for Calibrating Astrometry and Photometry); SExtractor generates object catalogs, while SCAMP conducts precision astrometric calibration, then MODP determines if a point source is moving. This package creates animated stamp images for visual inspection with MPC reports, the latter for checking whether an object is known or unknown. We evaluate the astrometric accuracy and efficiency of MODP using the year one dataset obtained from DEEP-South operations.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.4
• /
• pp.406-411
• /
• 2005

In this paper, we present the area-based Q-learning to search a target object using multiple robot. To search the target in Markovian space, the robots should recognize their surrounding at where they are located and generate some rules to act upon by themselves. Under area-based Q-learning, a robot, first of all, obtains 6-distances from itself to environment by infrared sensor which are hexagonally allocated around itself. Second, it calculates 6-areas with those distances then take an action, i.e., turn and move toward where the widest space will be guaranteed. After the action is taken, the value of Q will be updated by relative formula at the state. We set up an experimental environment with five small mobile robots, obstacles, and a target object, and tried to search for a target object while navigating in a unknown hallway where some obstacles were placed. In the end of this paper, we presents the results of three algorithms - a random search, area-based action making (ABAM), and hexagonal area-based Q-teaming.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.11
• /
• pp.2179-2184
• /
• 2007

The Intelligent Space(ISpace) provides challenging research fields for surveillance, human-computer interfacing, networked camera conferencing, industrial monitoring or service and training applications. ISpace is the space where many intelligent devices, such as computers and sensors, are distributed. According to the cooperation of many intelligent devices, the environment, it is very important that the system knows the location information to offer the useful services. In order to achieve these goals, we present a method for representing, tracking and human following by fusing distributed multiple vision systems in ISpace, with application to pedestrian tracking in a crowd. This paper described appearance based unknown object tracking with the distributed vision system in intelligent space. First, we discuss how object color information is obtained and how the color appearance based model is constructed from this data. Then, we discuss the global color model based on the local color information. The process of learning within global model and the experimental results are also presented.