• Journal of Advanced Marine Engineering and Technology
• /
• v.22 no.5
• /
• pp.652-659
• /
• 1998

The present study is aimed to develop a new PIV operating software through optimization of vector tracking identification including versatile pre-processings and post-processing techniques. And the result exhibits an improved version corresponding various input and output multimedia compared to previous commercial software developed by other makers. An upgraded identification method called grey-level cross correlation coefficient method by direct calculation is suggested and related user-friendly pop-up menu are also represented. Post-processings comprising turbulence statistics are also introduced with graphic output functions.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1996.04a
• /
• pp.556-560
• /
• 1996

In order to examine thd accuracy of intersection counting method, the fiber orientation distribution of simulation figure platted by PC is measured using image processing. The fiber orientation distribution obtained by an image processing method is compared with those by the intersection counting method. The result shows that the errors of the intersection counting method are large because its measurement is made by the cross point of the scanning line and the fiber.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2016.04a
• /
• pp.781-782
• /
• 2016

최근 증강현실 기술은 산업 분야뿐만 아니라 교육 분야에서도 많은 인기를 끌고 있다. 특히 유아 교육용 콘텐츠에서 특정 장면을 인식하고 장면에 해당하는 3차원 모형이 나타나는 형태는 아이의 호기심 자극하여 높은 교육효과를 나타내고 있다. 하지만 현재 증강현실 기술을 사용한 콘텐츠에서는 해당 장면에서 같은 내용만 나오기 때문에 반복적인 이용하는 경우에는 사용자가 지루함을 느끼는 문제점이 있다. 이를 보완하기 위해 사용자의 행위를 반영하는 증강현실 콘텐츠 제작하기 위하여 사용자의 행위가 반영된 마커 이미지를 실시간으로 추출하는 기법을 구현하고 실험하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.11a
• /
• pp.151-154
• /
• 2017

본 논문에서는 사용자의 주변사물을 인식하여 영단어로 알려줌으로써 사용자가 실생활에서 영단어를 능동적으로 학습할 수 있도록 돕는 애플리케이션을 개발한다. 본 애플리케이션은 사용자가 카메라로 촬영하거나 사진첩에서 선택한 이미지를 인식하여 사진 속 물체의 영어 단어와 한국어 뜻을 알려주며, 단어의 발음 또한 확인할 수 있고, 직접 단어장에 저장하여 다시 학습할 수 있도록 한다. 이를 위해 TensorFolw를 활용한 딥러닝 기반 이미지 인식 기술을 사용하였으며, 추후 TensorFolw를 통하여 모델을 추가적으로 훈련시킴으로써 이미지 인식의 정확도를 높일 수 있다. 그 외 영어-한국어 번역, 텍스트-음성 변환 등 부가 기능을 통해 사용자가 다양한 방식으로 영단어를 학습할 수 있도록 한다.

• Journal of Animal Environmental Science
• /
• v.6 no.1
• /
• pp.15-22
• /
• 2000

A monitoring and the forecasting system for the swine delivery was developed using CCD camera multi-function board microphone and data-recorder equipped on a personal computer. For the swine delivery monitoring and forecasting factors four factors were selected such as genitalia swine body shape breast color and sound. Image of physical variation of body shape, shape and color of genitalia area and color of breast of pregnant sow were grabbed using the CCD color camera and multi-function board and variation of voice of pregnant sow was acquired using microphone and data recorder. Acquired information of image and voice were analyzed using a custom developed algorithm and program. The result of the forecasting efficiency of swine delivery was 89%, 71% and 100% using the variation of genitalia are the body shape and the voice of pregnant sow. respectively. The efficiency of image processing was 100% for the delivery detection when the piglet was delivered half of its body from genitalia of pregnant sow, The monitoring and forecasting system informed the estimated time of the delivery of swine to a farm manager immediately if an estimated and established time set by the farm manager was the same and/or the estimated time ws earlier than the established time and the system detected the delivery.