• Journal of the Optical Society of Korea
• /
• v.20 no.2
• /
• pp.239-244
• /
• 2016

Mugshot face images, routinely collected by police, usually contain both frontal and profile views. Existing automated face recognition methods exploited mugshot databases by enlarging the gallery with synthetic multi-view face images generated from the mugshot face images. This paper, instead, proposes to match the query arbitrary view face image directly to the enrolled frontal and profile face images. During matching, the 3D face shape model reconstructed from the mugshot face images is used to establish corresponding semantic parts between query and gallery face images, based on which comparison is done. The final recognition result is obtained by fusing the matching results with frontal and profile face images. Compared with previous methods, the proposed method better utilizes mugshot databases without using synthetic face images that may have artifacts. Its effectiveness has been demonstrated on the Color FERET and CMU PIE databases.

• Journal of Korea Game Society
• /
• v.11 no.3
• /
• pp.3-8
• /
• 2011

The learning by computer games, is in the spotlight as an effective education method for the game generation. The educational games need to provide additionally an omniscient view of the learning as an objective viewpoint for storytelling of learning. The omniscient viewpoint needs good readability in real-time, and to provide systematically overall and detailed informations of learning. In this paper, we propose the visual learning map as the omniscient view for storytelling of learning. The proposed learning map is composed of the district map, the topic map, and the progress map for the omniscient view of the learning. The proposed learning map is represented by several visual diagrams for real-time readability. The proposed learning map can apply to all the educational games which provide the game minimaps.

• Journal of Broadcast Engineering
• /
• v.13 no.6
• /
• pp.915-929
• /
• 2008

In this paper, we propose an algorithm for pixel-based disparity estimation with reliability in the multi-view image. The proposed method estimates an initial disparity map using edge information of an image, and the initial disparity map is used for reducing the search range to estimate the disparity efficiently. Furthermore, disparity-mismatch on object boundaries and textureless-regions get reduced by adaptive block size. We generated intermediate-view images to evaluate the estimated disparity. Test results show that the proposed algorithm obtained $0.1{\sim}1.2dB$ enhanced PSNR(peak signal to noise ratio) compared to conventional block-based and pixel-based disparity estimation methods.

• Fashion & Textile Research Journal
• /
• v.13 no.1
• /
• pp.44-54
• /
• 2011

This study aims to clarify the asset evaluation attributes of imported brands exclusively by department stores based on qualitative study. For the qualitative research, hands-on staff of the branch office who had at least 5 years experience in PB-related positions at one of the four domestic department stores(Lotte, Shinsegae, Hyundai, Galleria), and at least 20 years old adults who had purchased imported brand items exclusively by a certain department store(Shinsegae, Lotte) were included. The asset evaluation attributes of department store-exclusively imported brands were drawn up through the one-to-one in-depth interview method. As a result of qualitative study on the asset evaluation attributes of department store-exclusive importation brands, it turned out that there were 4 factors in view of business entity - stability, profitability, growth potential, and relationship, and that there were 6 factors in view of consumers - product quality, recognition, country of origin, image, preference, and reliability. In comparison with the existing preliminary study, it was notable that relationship in view of business entity, and country of origin in view of consumers were added to the asset elements.

• The Korean Journal of Nuclear Medicine
• /
• v.20 no.1
• /
• pp.53-58
• /
• 1986

Thyroid scans using $Na^{131}I\;and\;^{99m}TcO_4^-$ in two (anterior and anterior oblique) projections were done simultaneously in 50 patients with single palpable nodule and in euthyroid state(nodular nontoxic goiter) and 20 normal subjects to evaluate the imaging ability of each radionuclide and the necessity of oblique view. In detection of a nodule as a cold or cool area, $^{99m}TcO_4^-$ was slightly superior to $^{131}I$(82 % to 78% with anterior view alone, 92% to 84% when oblique view is added in detectability), with addition of an oblique view, 5 more cold or cool area in $^{99m}Tc$ scan and 3 more $^{131}I$ scan were detected. In normal subjects $^{99m}TcO_4^-$ scan was significantly superior to $^{131}I$ scan(90% to 70%) in getting normal pattern of tracer uptake in the glands. For routine imaging study of the thyroid, $^{99m}Tc$ pertechnetate scan with multiple projection is considered to be a preferable method.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.11 no.5
• /
• pp.1558-1563
• /
• 2010

This paper describes a study on algorithm for the development of machine vision system as well as the inspection of automobile's plastic part locking lever to replace a human worker's eye inspection. Before developing the machine vision system based on the PC, the purpose of this research is to develop the algorithm to decide whether a product is a good/bad one in real time inspection. NI-LabVIEW software is used in the inspection method and an inspection program is developed using LabVIEW Vision image functions. The inspection program was built and validated to help the system operator set up the inspection area and change the criteria number in the program.

• The Journal of the Korea Contents Association
• /
• v.3 no.4
• /
• pp.17-22
• /
• 2003

In this paper, we propose the new side-view face detection method in color images which contain faces over one. It uses color and the geometrical distance between nose and chin. We convert RGB to YCbCr color space. We extract candidate regions of face using skin color information from image. And then, the extracted regions are processed by morphological filter, and the processed regions are labeled. Also, we correct the gradient of inclined face image using projected character of nose. And we detect the inclined side-view faces that have right and left 45 tips by within via ordinate. And we get 92% detection rate in 100 test images.

• Journal of KIISE:Databases
• /
• v.28 no.4
• /
• pp.620-632
• /
• 2001

This paper proposes a mediator system which can provide a view management function in the today's heterogeneous computing environment including operating system database management systems and other software. while most mediators use view modification mechanism only, the use of view materialization is also considered in this research. In order to optimize the management cost for materialized views the proposed method applies a decay factor for modeling the access behaviors of views In other words, the most recently access pattern gets the highest attention in the optimization process all the necessary it with the CORBA.

• Journal of the Korea Safety Management & Science
• /
• v.15 no.3
• /
• pp.61-69
• /
• 2013

According to ISO 26262 (the international standard on functional safety for automotive industry), the functional safety should be considered during the whole automotive systems life cycle from the design phase throughout the production phase. In order to satisfy the standard, the automotive and related industry needs to take appropriate actions while carrying out a variety of development activities. This paper presents an approach to coping with the standard. Analyzing the standard indicates that the safety issues of the automotive systems should be handled with a system's view whereas the conventional approach to solving the issues has been practiced with focus on the component's level. The aforementioned system's view implies that the functional safety shall be incorporated in the system design from both the system's life-cycle view and the hierarchical view for the structure. In light of this, the systems engineering framework can be quite appropriate in the functional safety development and thus has been taken in this paper as a problem solving approach. Of various design issues, the analysis and verification of the safety requirements for functional safety is a key study subject of the paper. Note, in particular, that the conventional FMEA (failure mode effects analysis) and FTA (fault tree analysis) methods seem to be partly relying on the insufficient experience and knowledge of the engineers. To improve this, a systematic method is studied here and the result is applied in the design of an ABS braking system as a case study.

• Journal of Broadcast Engineering
• /
• v.21 no.1
• /
• pp.66-75
• /
• 2016

In this paper, we present a new method to reduce crosstalk of a glasses-free three-dimensional (3D) display using a multi-view image processing technique. Since crosstalk makes the current view image mixed with its neighboring ones, the output 3D image becomes severely blurred. We apply adaptive depth retargeting and view gradient-based crosstalk inverse filtering onto a multi-view image to minimize crosstalk of the glasses-free 3D display. In addition, overflow and underflow pixels are compensated by epipolar image pixel interpolation so that visual artifacts are minimized. Experimental results show that the proposed method reduces crosstalk more efficiently than the previous work while improving 3D image quality.