• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.12
• /
• pp.2547-2554
• /
• 2009

In this paper, we proposed a face detection algorithm and a hardware implementation method for ROI(Region Of Interest) of AF(Auto Focusing). We used face features in skin regions of YCbCr color space for face detection. The face features are the number of skin pixels in face regions, edge pixels in eye regions, and shadow pixels in lip regions. The each feature was statistically selected by 2,000 sample pictures of face. The proposed algorithm detects two faces that are closer center of the image for considering the effectiveness of hardware resource. The detected faces are displayed by rectangle for ROI of AF, and the rectangles are represented by positions in the image about starting point and ending point of the rectangles. The proposed face detection method was verified by using FPGA boards and mobile phone camera sensor.

• Korean Institute of Interior Design Journal
• /
• v.19 no.1
• /
• pp.180-189
• /
• 2010

Since collective houses are started to be supplied, the housing market has been transformed from supplier to demander oriented, and various type with same area of unit households have been deigned and started to be distributed. It can be expected that such transfer phenomenon would be continuously sped-up, and unit household will be more subdivided and varied. Thus, we analysed and studied the planes with subject of unit households in national housing scale ($85m^2$) that is treated as supply standard and has biggest volume in construction by Korean housing corporation. The analysing method is investigating the area of each room within unit household, calculating their minimum, maximum, and average values, studying the efficiency of unit household plane according to the structure of each room by comparative analysis these values, and analysing changing matter by investigating any change of each room area. The analysing results of which are followings. At first, number of BAY on face is increasing. Second, allocation of each room has been vary. Third, area ratio among bedroom1, livingroom, diningroom/kitchen become similar. Fourth, the area of diningroom/kitchen become somewhat bigger. Fifth, the space for passage way is efficiently allocated. Sixth, service area is reduced, Seventh, stereotyped phenomenon Is appeared. Based on many studies so far, any improvements of unit householder in national housing scale are followings. First, in terms of forming style, unit householder in national housing scale is unfolded in various way though, the form of each room is not more than typical style. Second, it shows stereotyped phenomenon. Especially, bathroom 1, 2 show stereotyped feature in terms of their type and area across entire research years, also LDK allocation method is stereotyped, as well as stereotyped trend in bedroom 2, 3, allocation for 4 BAY type.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.10 no.6
• /
• pp.594-600
• /
• 2017

The new design methodology of prediction model and pattern classification, which is based on the dimension reduction algorithm called principal component analysis, is introduced in this paper. Principal component analysis is one of dimension reduction techniques which are used to reduce the dimension of the input space and extract some good features from the original input variables. The extracted input variables are applied to the prediction model and pattern classifier as the input variables. The introduced prediction model and pattern classifier are based on the very simple regression which is the key point of the paper. The structural simplicity of the prediction model and pattern classifier leads to reducing the over-fitting problem. In order to validate the proposed prediction model and pattern classifier, several machine learning data sets are used.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.27 no.5
• /
• pp.1077-1086
• /
• 2017

Malware mutants based on existing malware is widely used because it can easily avoid the existing security system even with a slight pattern change. These malware appear on average more than 1.6 million times a day, and they are gradually expanding to IoT / ICS as well as cyber space, which has a large scale of damage. In this paper, we propose an analytical method based on features of PE Section and DLL that do not give much significance, rather than pattern-based analysis, Sandbox-based analysis, and CFG, Strings-based analysis. It is expected that the proposed model will be able to cope with effective malicious code in case of combined operation of various existing analysis technologies.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.5
• /
• pp.3619-3626
• /
• 2015

The coal shed at thermal power plant(T.P.P) is the biggest building on size among nuclear, coal and other power plant industry buildings. This the coal shed of indoor type is mass volume, the size of 6 soccer fields and 73 meter in height. Structural and functional elements take priority over aesthetic factors. The shed is built to make mighty space for structural safety and functional store by using the concrete, $75,000m^3$ on the total of central retaining wall and sub buttress, and the steel frame used by 11,744-ton P.E.B. system. The design requirement on its wall and frame, P.E.B. system's feature, and specific data of main process followed by construction are inquired. The aim of this study is to suggest the design requirement and guide for the indoor type of the coal shed.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.11 no.3
• /
• pp.847-852
• /
• 2010

Text in natural images has a various and important feature of image. Therefore, to detect text and extraction of text, recognizing it is a studied as an important research area. Lately, many applications of various fields is being developed based on mobile phone camera technology. Detecting edge component form gray-scale image and detect an boundary of text regions by local standard deviation and get an connected components using Euclidean distance of RGB color space. Labeling the detected edges and connected component and get bounding boxes each regions. Candidate of text achieved with heuristic rule of text. Detected candidate text regions was merged for generation for one candidate text region, then text region detected with verifying candidate text region using ectilarity characterization of adjacency and ectilarity between candidate text regions. Experctental results, We improved text region detection rate using completentary of edge and color connected component.

• The KIPS Transactions:PartB
• /
• v.10B no.1
• /
• pp.47-56
• /
• 2003

Korean delimits words by white-space like English, but words In Korean Is a little different in structure from those in English. Words in English generally consist of one word, but those in Korean are composed of one word and/or morpheme or more. Because of this difference, a word between white-spaces is called an Eojeol in Korean. We propose a method for segmenting and classifying Korean words and/or morphemes based on syllables using an instance-based learning. In this paper, elements of feature sets for the instance-based learning are one previous syllable, one current syllable, two next syllables, a final consonant of the current syllable, and two previous categories. Our method shows more than 97% of the F-measure of word segmentation using ETRI corpus and KAIST corpus.

• Journal of Internet Computing and Services
• /
• v.5 no.6
• /
• pp.79-91
• /
• 2004

Various facial pose detection and recognition has been a difficult problem. The problem is due to the fact that the distribution of various poses in a feature space is mere dispersed and more complicated than that of frontal faces, This thesis proposes a robust pose-expression-invariant face recognition method in order to overcome insufficiency of the existing face recognition system. First, we apply the TSL color model for detecting facial region and estimate the direction of face using facial features. The estimated pose vector is decomposed into X-V-Z axes, Second, the input face is mapped by deformable template using this vectors and 3D CANDIDE face model. Final. the mapped face is transformed to frontal face which appropriates for face recognition by the estimated pose vector. Through the experiments, we come to validate the application of face detection model and the method for estimating facial poses, Moreover, the tests show that recognition rate is greatly boosted through the normalization of the poses and expressions.

• Journal of Internet Computing and Services
• /
• v.5 no.2
• /
• pp.97-105
• /
• 2004

The reason that an intelligent agent cannot understand the representation of its own perception or activity is caused by the traditional syntactic approach that translates a semantic feature into a simulated string, To implement an autonomously learning intelligent agent, Cohen introduces a experimentally semantic approach that the system learns a contentful representation of physical schema from physically interacting with environment using its own sensors and effectors. We propose that negation is a meta-level schema that enables an agent to recognize its own physical schema, To improve the planner's efficiency, Graphplan introduces the control rule that manipulates the inconsistency between planning operators, but it cannot cognitively understand negation and suffers from redundancy problem. By introducing a negative function not, IPP solves the problem, but its approach is still syntactic and is inefficient in terms of time and space. In this paper, we propose that, to represent a negative fact, a positive atom, which is called opposite concept, is a very efficient technique for implementing an cognitive agent, and demonstrate some empirical results supporting the hypothesis.

• Journal of Fashion Business
• /
• v.20 no.4
• /
• pp.15-35
• /
• 2016

The purpose of this study was to develop brassiere pattern designed to fit the breast shapes based on ESMOD pattern. It has three quarters cup round shape and also consists of three parts; upper cup, lower cup, and wings. Breast types are classified into five shapes; ideal breast, flat breast, upper developed breast, lower developed breast, and projecting breast. Two subjects for each breast type wore the brassiere, and they evaluated the appearance and wearing twice. Type I for research pattern designed to fit into the breast shape reflecting details of breast size were assessed as superior to the divided commercial type. However, wings' tightness of Type I for research pattern brassiere was high. Thus, to improve wearing satisfaction, extra was added to wing. Based on the results of wearing experiments of Type I for research, we adjusted and modified Type II for research pattern. Subsequently, its appearance and wearing were evaluated, in order to be improved. For upper developed breast pattern, we extended the length of lower part to balance upper and lower part, as the upper part was somewhat long. The lower developed breast has the closest feature to the ideal breast, suggestive that implies it does not require much improvement Projecting breast pattern has minimal space in the lower part, so we added the support to lift them to be similar to the ideal breasts. For all the breast shapes, we reduced the wings' tightness from 8% to 7% so that we could extend the length of the wings.