• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.701-706
• /
• 2009

We proposed the device to control a computer with only a head and eye blinks so that disabilities by car accidents can use a computer. Because they have paralysis of their upper extremities such as C4~C5 paraplegics and cerebral palsy, they cannot efficiently access a general keyboard/mouse not using hands and foots. The cursor position was estimated from a gyro-sensor which can measure head movements, and the mouse event such as click/double click from opto-sensors which can detect eye blinks. The sensor was put on the proper goggle in order not to disturb the visual field. The performance of the proposed device was compared to a general optical mouse, and was used both relative and absolute coordinate in cursor positioning control. The recognition rate of click and double-click was 86% of the optical mouse, the speed of cursor movement by the proposed device was not much different from the mouse. The overall accuracy was 80%. Especially, the relative coordinate is more convenience and accuracy than the absolute coordinate, and can reduce the frequency of reset to prevent the accumulative error.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.4
• /
• pp.60-68
• /
• 1994

In this paper we propose a new Hybrid learning rule applied to multilayer feedforward neural networks, which is constructed by combining Hebbian learning rule that is a good feature extractor and Back-Propagation(BP) learning rule that is an excellent classifier. Unlike the BP rule used in multi-layer perceptron(MLP), the proposed Hybrid learning rule is used for uptate of all connection weights except for output connection weigths becase the Hebbian learning in output layer does not guarantee learning convergence. To evaluate the performance, the proposed hybrid rule is applied to classifier problems in two dimensional space and shows better performance than the one applied only by the BP rule. In terms of learning speed the proposed rule converges faster than the conventional BP. For example, the learning of the proposed Hybrid can be done in 2/10 of the iterations that are required for BP, while the recognition rate of the proposed Hybrid is improved by about $0.778\%$ at the peak.

• Journal of KIISE
• /
• v.41 no.11
• /
• pp.921-926
• /
• 2014

Face detection refers to the process extracting facial regions in an input image, which can improve speed and accuracy of recognition or authorization system, and has diverse applicability. Since conventional works have tried to detect faces based on the whole shape of faces, its detection performance can be degraded by occlusion made with accessories or parts of body. In this paper we propose a method combining local feature descriptors and probability modeling in order to detect partially occluded face effectively. In training stage, we represent an image as a set of local feature descriptors and estimate a statistical model for normal faces. When the test image is given, we find a region that is most similar to face using our face model constructed in training stage. According to experimental results with benchmark data set, we confirmed the effect of proposed method on detecting partially occluded face.

• Journal of Navigation and Port Research
• /
• v.32 no.5
• /
• pp.333-339
• /
• 2008

When a vessel is underway in a heavy weather, passengers and crew suffer from seasickness caused by ship motions such as pitch, heave or roll, or all combined Sickness induces drowsiness, dizziness, headache, stomachache etc, in extreme conditions, they are met with a serious trouble which is physiologically unrecoverable. It results in weakening of spiritual activities or making errors from decrease of motivation, dropping off skills, poor recognition and poor judgement. In this paper, it was examined the international standards concerning the occurrence of sickness and the execution of works, also evaluated the boarding comfort by conducting several times of questionnaire on cadets boarding on the training ship HANBADA As a result, it was confirmed that the main factor of occurring the sickness was the vertical acceleration and the level was more than 0.2g. Also, it was presented the way how to reduce the sickness by changing the speed and/or course in relation to the encounter period.

• Proceedings of the Korea Technical Association of the Pulp and Paper Industry Conference
• /
• 2007.05a
• /
• pp.33-55
• /
• 2007

This article portrays special cellulose fibers, which are designed to be a functional additive and a carrier for papermaking chemicals. The first part of the presentation deals with the micronized $ARBOCEL^{(R)}$ cellulose fibers, which are used as a functional paper/paperboard additive. In particular as a bulk and speed aid. The detailed description of the micronized $ARBOCEL^{(R)}$ fibers, their function and effects on papermaking process and paper products are given. The second part of the study describes the concept of fiber-based papermaking chemicals. A new generation of fiber-based papermaking chemicals were presented for the first time at the PTS Pulp Technology Symposium 2005, and then several articles were published in various magazine in Asia ("Paper Asia"), the US ("Pulp & Paper"). and Europe ("Wochenblatt fuel Papierfabrikation"). The information generated quite an interest in the paper industry. Extensive studies of these papermaking additives have been made recently, new information obtained, and the compounds have gained more recognition in the industry. The company J. Rettenmaier und Soehne developed a group of fiber-based papermaking additives. They include combination of fibers with sizing agents, starch, fluorochemicals, minerals, biocides and some others. This article presents in-depth study of the AKD modified micronized cellulose as an example of the fiber-based papermaking chemicals concept. The material of the present paper is based mostly on the results of the pilot paper machine study at the Paper Research Institute PTS (Heidenau, Germany), and includes case studies from the mills, which used $ARBOCELPLUS^{(R)}-AKD$ compounds. It should be noted that the $ARBOCELPLUS^{(R)}$ compounds were not designed to replace traditional additives in paper industry. They should rather be used in those areas, where application of "normal" chemicals is especially problematic

• Korean Institute of Interior Design Journal
• /
• v.14 no.6 s.53
• /
• pp.120-131
• /
• 2005

Nowadays, food style would be changing as western and much like simple. because there was changing for customer both life style and recognition of what the food is, so that the customer would like to get something that has a high qualified and special. In the society changing continuary, the upgraded speed of atmospher and competition between bakery company is running quickly. in order to follow the customer who are different individually and pursuit lots of kind of things. In fact, it's not only making atmosphere that could give relaxation and rich life but also supporting commodity and many kinds of information to effect the sale's effectiveness, so the various alternative need to be checked out to treat the difference of atmosphere. importance of space composition plan considerating specific character is confirmed each commercial area of alternatives. In this study, we want to provide the most important standard of space composition of several visible or invisible environment factor making an bakery store. commercial rights are defferentiated between center commercial land and dwelling commercial land and we can strengthen the competitveness by bringing inspect, analysis, comparison about product make-up price each commercial land in space organixation of in-store actively in the 21 century's new bakery competition. In the 21st century, we should recognize we are in the new period about these a aggressive goal of maximization of sales and performed space construction through plan specialization to build reasonably and effectively. composition plan for store space as composition rate of product would be basic plan. so when the opening store is decided it need to be basic composition plan space which Is proper to situation of area.

• Korean Institute of Interior Design Journal
• /
• v.23 no.5
• /
• pp.104-111
• /
• 2014

This study puts its purpose on presenting an integral system of representative urban landscapes, public design and colors, in practicing urban design in the context of local identity. An integral design of public design that successfully plans and executes local identity, and changes recognition of integral management of urban design, is suggested as followings. Firstly, when the catchment area is divided in metropolitan area, it should be reset according to the natural environment condition regardless of administrative area system like city and county. It is the method to classify the metropolitan area by researching and analyzing geographical condition, weather condition, soil and vegetation in detail and subclassify it by the visual commonness of natural environment. Secondly, it is necessary to access the urban landscape, public design and urban color from the overall aspect emphasizing the plan for each field and local identity. They should be practiced by the role and category of each field on the basis of consistent design strategy and instruction but the cooperation system is required as a process to reinforce and specify the mutual limit. Thirdly, the artificial structure is constructed through artificial adjustment depending on the urban formation process and the development time point. Therefore, it is necessary to pay attention to the rapid urban development, the change speed and the landscape formation of each age. It is necessary to classify the type of artificial landscape by age and form similarity and separate the area that should be generalized and controlled by entire metropolitan area form the area that should be specialized by basic local government.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.4
• /
• pp.48-53
• /
• 2001

In this paper, selectively attentive learning method has been proposed to improve the learning speed of multilayer Perceptron based on the error backpropagation algorithm. Three attention criterions are introduced to effectively determine which set of input patterns is or which portion of network is attended to for effective learning. Such criterions are based on the mean square error function of the output layer and class-selective relevance of the hidden nodes. The acceleration of learning time is achieved by lowering the computational cost per iteration. Effectiveness of the proposed method is demonstrated in a speaker adaptation task of isolated word recognition system. The experimental results show that the proposed selective attention technique can reduce the learning time more than 60% in an average sense.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.2568-2574
• /
• 2005

Control architecture of the action based robot engineering can be divided into two types of deliberate type - and reactive type- controller. Typical deliberate type, slow in reaction speed, is well suited for the realization of the higher intelligence with its capability to forecast on the basis of environmental model according to time flow, while reactive type is suitable for the lower intelligence as it fits to the realization of speedy reactive action by inputting the sensor without a complete environmental model. Looking at the environments in the application areas in which robots are actually used, we can see that they have been mostly covered by the uncertain and unknown dynamic changes depending on time and place, the previously known knowledge being existed though. It may cause, therefore, any deterioration of the robot performance as well as further happen such cases as the robots can not carry out their desired performances, when any one of these two types is solely engaged. Accordingly this paper aims at suggesting Goal-oriented Geometric Model(GGM) Based Intelligent System Architecture which leads the actions of the robots to perform their jobs under variously changing environment and applying the suggested system structure to the navigation issues of the robots. When the robots do perform navigation in human life changing in a various manner with time, they can appropriately respond to the changing environment by doing the action with the recognition of the state. Extending this concept to cover the highest hierarchy without sticking only to the actions of the robots can lead us to apply to the algorithm to perform various small jobs required for the carrying-out of a large main job.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.9
• /
• pp.75-81
• /
• 2014

The integral image is the sum of input image pixel values. It is mainly used to speed up processing of a box filter operation, such as Haar-like features. However, large memory for integral image data can be an obstacle on an embedded hardware environment with limited memory resources. Therefore, an efficient method to store the integral image is necessary. In this paper, we propose a memory size reduction hardware design for integral image. The hardware design is used two methods. It is the new integral image memory and modulo calculation for reducing integral image data. The new integral image memory has additional calculation overhead, but it is not obstacle in hardware environment that parallel processing is possible. In the Xilinx Virtex5-LX330T targeted experimental result, integral image memory can be reduced by 50% on a $640{\times}480$ 8-bit gray-scale input image.