• Annual Conference of KIPS
• /
• 2013.11a
• /
• pp.1439-1441
• /
• 2013

본 연구에서는 비디오 감시 시스템을 기반으로 한우 축사에서 승가 행위 검출을 위한 최적의 카메라 위치를 결정한다. 실외 환경에서는 소들간의 겹침이나 조명 변화 등 다양한 어려움이 발생하기 때문에, 이를 극복하기 위하여 승가 시 소의 몸체가 올라간다는 특성을 이용한다. 즉, 등높이가 1.2m에서 1.3m 사이 크기의 한우를 사육하는 축사에서 축사 측면에 1.55m 높이로 카메라를 설치하고 여기서 획득된 영상으로 실험한 결과, 발정기 탐지를 위한 승가 행위를 자동으로 검출할 수 있음을 확인하였다.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.198-204
• /
• 2022

대화 요약은 다중 발화자와 발화문으로 이루어진 멀티턴 형식의 문서에 대해 핵심내용을 추출하거나 생성하는 태스크이다. 대화 요약 모델은 추천, 대화 시스템 등에 콘텐츠, 서비스 기록에 대한 분석을 제공하는 데 유용하다. 하지만 모델 구축에 필요한 한국어 대화 요약 데이터셋에 대한 연구는 부족한 실정이다. 본 논문에서는 생성 기반 대화 요약을 위한 데이터셋을 제안한다. 이를 위해 국내 방송사의 대용량 콘텐츠로 부터 원천 데이터를 수집하고, 주석자가 수작업으로 레이블링 하였다. 구축된 데이터셋 규모는 6개 카테고리에 대해 약 100K이며, 요약문은 단문장, 세문장, 2할문장으로 구분되어 레이블링 되었다. 또한 본 논문에서는 데이터의 특성을 내재화하고 통제할 수 있도록 대화 요약 레이블링 가이드를 제안한다. 이를 기준으로 모델 적합성 검증에 사용될 디코딩 모델 구조를 선정한다. 실험을 통해 구축된 데이터의 몇가지 특성을 조명하고, 후속 연구를 위한 벤치마크 성능을 제시한다. 데이터와 모델은 aihub.or.kr에 배포 되었다.

• Annual Conference of KIPS
• /
• 2021.11a
• /
• pp.290-292
• /
• 2021

양자 컴퓨터 기술의 발전에 따라 현재 사용되고 있는 암호 알고리즘과 시스템들이 위협받고 있다. 이러한 시대적 흐름에 따라 양자 컴퓨터로도 쉽게 해결할 수 없는 양자내성암호의 개발이 요구되고 있으며, 미국 NIST 에서는 양자내성암호의 표준화를 위한 공모전을 진행하고 있다. 본 논문에서는 공모전 최종 후보 중 하나인 NTRU 알고리즘을 가속화하기 위한 성능 분석을 진행하였다.

• Journal of the Korean Society for Precision Engineering
• /
• v.12 no.10
• /
• pp.121-129
• /
• 1995

The objective of this research is to develop an effective inspection planning strategy for sculptured surfaces by using 3-dimensional Coordinate Measuring Machine (CMM). First, the CAD/CAM database is generated by using the Bezier surface patch mathod and variable cutter step size approach for design and machining of the workpiece model. Then, optimum measuring point locations are determained based on the mean curvature analysis to obtain more effective inspection results for the given sample numbers. An optimal probe sequence generation method is proposed by implementing the Traveling Salesperson (TSP) algorithm and new guide point selection methods are suggested based on the concepts of the variable distance between the first and second guide points. Finally, simulation study and experimental work show the effectiveness of the proposed strategy.

• Annual Conference of KIPS
• /
• 2008.11a
• /
• pp.149-152
• /
• 2008

사생활 보호에 대한 인식이 커지고, 인터넷 시대에 접어들면서 네트워크 기반의 보안시스템의 개발이 활발하다. 실시간 비디오 카메라를 통한 움직이는 물체를 검출하기 위해서는 불필요한 잡음이나 조명의 변화에 대처해야 한다. 이러한 많은 요소들을 고려하여 움직이는 물체를 검출하려면 많은 계산 복잡도를 가지게 된다. 또한, 카메라의 영상크기가 증가함에 따라 움직이는 물체를 검출하기 위해서 더 많은 계산 복잡도를 가지게 된다. 본 논문에서는 기존의 통상적인 움직임 검출방법 과 적응적 배경방식인 '물체 검출을 위한 동적인 장면의 베이시안 모델링 기반 물체 검출 방법'을 분석하고, 실시간으로 처리되는 동적 비디오 영상에서 이동 물체를 검출하는 과정에서의 영상의 크기가 커지고, 이동하는 물체의 개수가 많아짐에 따라 발생되는 계산의 복잡도를 'CPU 성능과 영상 resize 를 이용한 계산 복잡도 감소 방법'을 통해 초당 프레임 처리속도를 유지시키는 방법을 제시한다.

• Journal of KIISE:Software and Applications
• /
• v.30 no.9
• /
• pp.829-842
• /
• 2003

In this Paper, we propose the new face detection and tracking method based on template matching for real-time applications such as, teleconference, telecommunication, front stage of surveillance system using face recognition, and video-phone applications. Since the main purpose of paper is to track a face regardless of various environments, we use template-based face tracking method. To generate robust face templates, we apply wavelet transform to the average face image and extract three types of wavelet template from transformed low-resolution average face. However template matching is generally sensitive to the change of illumination conditions, we apply Min-max normalization with histogram equalization according to the variation of intensity. Tracking method is also applied to reduce the computation time and predict precise face candidate region. Finally, facial components are also detected and from the relative distance of two eyes, we estimate the size of facial ellipse.

• 한국산학경영학회:학술대회논문집
• /
• 2005.11a
• /
• pp.29-46
• /
• 2005

Recently, TPS is one of the hot issues which enterprises and manufacturing related people were greatly concerned. TPS is very representative of Japanese production system. And TPS have contributed to the advance of industrial engineering. This study highlights the limitation and unsolved problems of presented TPS, through understanding the basic principles and characteristics and application process of TPS. To be continuously developing and competitive system in the future, we suggest New JIT System to complement current TPS system. New JIT is a customer oriented system. For the customer satisfied quality management, renovate business process in production, development and sales area. It has hardware and software system. Hardware system is consists of TMS, TDS, TPS. Software system Is scientific SQC utilized TQM. For the advanced next generation manufacturing management system, we should develop New JIT system.

• Journal of Digital Contents Society
• /
• v.7 no.4
• /
• pp.287-294
• /
• 2006

Mass-customization is fast growing a segment of the apparel market. 2D Virtual wearing system is one of visual support tools that make possible to sell apparel before producing and reduce the time and costs related to product development and manufacturing in the world of apparel mass-customization. This paper is related to fabric color mapping method for 2D image-based virtual wearing system. In proposed method, clothing shape section of interest is segmented from a clothes model image using a region growing method, and then mapping a new fabric color selected by user into it based on its intensity difference map is processed. With the proposed method in 2D virtual wearing system, regardless of color or intensity of model clothes, it is possible to virtually change the fabric color with holding the illumination and shading properties of the selected clothing shape section, and also to quickly and easily simulate, compare, and select multiple fabric color combinations for individual styles or entire outfits.

• Journal of Korea Multimedia Society
• /
• v.13 no.4
• /
• pp.504-511
• /
• 2010

In this paper, we propose the face region recognition system for City-Security CCTV(Closed Circuit Television) using hippocampal neural network which is modelling of human brain's hippocampus. This system is composed of feature extraction, learning and recognition part. The feature extraction part is constructed using PCA(Principal Component Analysis) and LDA(Linear Discriminants Analysis). In the learning part, it can label the features of the image-data which are inputted according to the order of hippocampal neuron structure to reaction-pattern according to the adjustment of a good impression in a dentate gyrus and remove the noise through the auto-associative memory in the CA3 region. In the CA1 region receiving the information of the CA3, it can make long-term memory learned by neuron. Experiments confirm the each recognition rate, that are shape change and light change. The experimental results show that we can compare a feature extraction and learning method proposed in this paper of any other methods, and we can confirm that the proposed method is superior to existing methods.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.3C
• /
• pp.125-132
• /
• 2005

This paper describes a computer vision system based on active IR illumination for real-time gaze discrimination system. Unlike most of the existing gaze discrimination techniques, which often require assuming a static head to work well and require a cumbersome calibration process for each person, our gaze discrimination system can perform robust and accurate gaze estimation without calibration and under rather significant head movement. This is made possible by a new gaze calibration procedure that identifies the mapping from pupil parameters to screen coordinates using generalized regression neural networks (GRNNs). With GRNNs, the mapping does not have to be an analytical function and head movement is explicitly accounted for by the gaze mapping function. Futhermore, the mapping function can generalize to other individuals not used in the training. To further improve the gaze estimation accuracy, we employ a reclassification scheme that deals with the classes that tend to be misclassified. This leads to a 10% improvement in classification error. The angular gaze accuracy is about 5°horizontally and 8°vertically. The effectiveness of our gaze tracker is demonstrated by experiments that involve gaze-contingent interactive graphic display.