• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.25 no.6
• /
• pp.1485-1494
• /
• 2015

Traditionally, data hiding has been done mainly in such a way that insert the data into the large-capacity multimedia files. However, the document files of the previous versions of Microsoft Office 2003 have been used as cover files as their structure are so similar to a File System that it is easy to hide data in them. If you open a compound-document file which has a secret message hidden in it with MS Office application, it is hard for users who don't know whether a secret message is hidden in the compound-document file to detect the secret message. This paper presents an analysis of Compound-File Binary Format features exploited in order to hide data and algorithms to detect the data hidden with these exploits. Studying methods used to hide data in unused area, unallocated area, reserved area and inserted streams led us to develop an algorithm to aid in the detection and examination of hidden data.

• Convergence Security Journal
• /
• v.23 no.5
• /
• pp.117-123
• /
• 2023

Recently, as the demand for Generative Artificial Intelligence (AI) and artificial intelligence has increased, the seriousness of misuse and abuse has emerged. However, intelligent CCTV, which maximizes detection of abnormal behavior, is of great help to prevent crime in the military and police. AI performs learning as taught by humans and then proceeds with self-learning. Since AI makes judgments according to the learned results, it is necessary to clearly understand the characteristics of learning. However, it is often difficult to visually judge strange and abnormal behaviors that are ambiguous even for humans to judge. It is very difficult to learn this with the eyes of artificial intelligence, and the result of learning is very many False Positive, False Negative, and True Negative. In response, this paper presented standards and methods for clarifying the learning of AI's strange and abnormal behaviors, and presented learning measures to maximize the judgment ability of intelligent CCTV's False Positive, False Negative, and True Negative. Through this paper, it is expected that the artificial intelligence engine performance of intelligent CCTV currently in use can be maximized, and the ratio of False Positive and False Negative can be minimized..

• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 2002.05a
• /
• pp.90-94
• /
• 2002

이전연구에서 가상현실에 몰입하는 동안 넓은 시야 (Field of view: 150$^{\circ}$)와 빠른 운행속도 (70 km/sec)가 사이버멀미를 심화시킨다는 결과를 얻었다: 피험자의 90%가 좁은 시야 (50$^{\circ}$)와 느린 운행속도 (30 km/sec)에서 사이버멀미 증상이 적었다. 본 실험에서는 피험자가 생리적인 동요를 경험할 때마다 바이오피드백 방법을 사용해서 사이버멀미 감소 가상환경 (cybersickness alleviating virtual environment, CAVE)을 제시한 후, 그 효과를 관찰하였다. 피부전도도, 말초체온, 말초혈류량, 심박률, 눈 깜박임, 뇌전위의 변수들을 입력으로 하는 인공신경망으로 구성된 실시간 멀미 탐지 시스템과 CAVE-제시 피드백 시스템을 구축하였다. 이 시스템은 생리적 측정치들이 사이버멀미의 출현을 신호할 때마다 피드백 출력으로 좁은 화면과 감소된 운행속도를 일시적으로 제공했다. 36명의 피험자를 대상으로 SSQ (simulator sickness questionnaires)와 자기보고를 이용하여 사이버멀미의 빈도와 심각도를 조사하였다. 모든 피험자는 한달 간격으로 CAVE 조건과 non-CAVE 조건에서 두 번 가상현실을 경험하였다. 사이버멀미의 빈도와 심각도는 non-CAVE 조건보다 CAVE 조건에서 유의미하게 감소하였다. 즉, 전기생리학적 특징들에 기반한 인공신경망에 의해 제공된 좁은 시야와 느린 운행의 가상환경은 사이버멀미 증상들을 의미있게 감소시켰다. 이러한 결과들은 생체신호 피드백 시스템을 이용하여 인간 친화적 가상환경을 구축할 수 있는 가능성을 보인 것이다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.1A
• /
• pp.89-97
• /
• 2002

This paper proposes the adaptive digital watermarking method for the ownership protection of images. The watermarks are inserted to a selected area rather than a whole area. The proposed method reduces the distortion caused by the watermarking process. To select the regions, roughness of the image should be considered because the watermarks in the smooth regions are easily detected through the human eyes. To find the rough regions, Discrete Cosine Transform (DCT) method is used. Generally, the high frequency regions of images are lost by the compression process such as JPEG. So, the watermarks are inserted to the low frequency regions of a selected area by using the proposed method. The proposed method reduce the image loss or distortion brought by the image processing, such as compression, filtering, scaling, addition of noise, cropping, and wavelet transform.

• Journal of Convergence Society for SMB
• /
• v.3 no.2
• /
• pp.19-25
• /
• 2013

Decision procedure was done with the evaluation of multi-criterion analysis. Importance of each criterion was considered through heuristically method, specially it was based on the heuristic least mean square algorithm. To consider coalition evaluation, it was carried out by calculation of Shapley index and Interaction value. The model output is also analyzed with the help of those two indexes, and the procedure was also displayed with details. Finally, the differences between the model output and the desired results are evaluated thoroughly, several problems are raised at the end of the example which require for further studying.

• Journal of Convergence Society for SMB
• /
• v.3 no.1
• /
• pp.31-41
• /
• 2013

There have been various recent methods proposed in detecting driver drowsiness (DD) to avert fatal accidents. This work proposes a hardware/software (HW/SW) co-design approach in implementation of a DD detection system adapted from an AdaBoost-based object detection algorithm with Haar-like features [1] to monitor driver's eye closure rate. In this work, critical functions of the DD detection algorithm is accelerated through custom hardware components in order to speed up processing, while the software component implements the overall control and logical operations to achieve the complete functionality required of the DD detection algorithm. The HW/SW architecture was implemented on an Altera DE2 board with a video daughter board. Performance of the proposed implementation was evaluated and benchmarked against some recent works.

• Proceedings of the KSRS Conference
• /
• 2009.03a
• /
• pp.105-110
• /
• 2009

현재 NASA에서 제공되는 MODIS 지표반사율자료(MOD09)는 MODIS영상을 이용한 각종 주제자료들의 중요한 입력 자료로 사용되고 있으며, MODIS 지표반사율 자료에 대한 객관적인 검증연구가 필요한 실정이다. 따라서 본 연구에서는 MOD09의 검증관련 초기 연구로서, 남한에 분포하는 불변성 타겟(invariant target)을 대상으로 2006년 일별 250m MODIS 지표반사율자료(MOD09GQK)자료의 객관적 검증을 시도하였다. 우선, MOD09 QA(Quality Assurance)자료를 이용하여 구름의 영향을 받은 화소를 제거한 후, 수치지도와 토지피복도를 이용하여 정의한 불변성 타겟에 해당되는 MOD09영상의 화소값을 추출하였다. 이와 같이 추출된 시계열 MOD09GHK영상의 화소값에 1차 회귀분석을 적용하여 이상 반사율 값을 탐지하고, 그 원인을 분석하였다. 검증 결과 나지지역에 대해서 0.0186의 RMSE값이 나타났으며, 인공물의 경우 0.2891의 RMSE값을 보였다. 발생된 이상 화소를 살펴보면, 구름, 그림자, 눈에 영향에 의해 발생한 것도 있으며, 원인을 알 수 없는 이상 화소들도 분포하였다. 향후 연구에서는 한반도 전역의 MODIS 시계열 반사율영상을 대상으로 MODIS 대기보정알고리즘과 입력인자의 적합성을 판단하기 위한 연구를 진행할 예정이다.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.397-402
• /
• 2022

익명 네트워크 기술에 기반한 다크웹은 일반 표면웹보다 더 강화된 익명성을 제공한다. 최근 이 익명성을 악용하여 다수의 다크웹 사용자들이 다크웹 내에서 범죄 행위를 모의하는 행위가 꾸준히 발생하고 있다. 특히, 국내 다크웹 사용자들은 마약 유포를 위한 방법을 공유하거나 성착취물 유포 행위 등에 직간접적으로 가담하고 있다. 이와 같은 범죄 행위들은 수사 기관의 눈을 피해 현재까지도 계속해서 발생하고 있어 국내 다크웹 범죄 동향 파악의 필요성이 증대되고 있다. 그러나 다크웹 특성상 범죄 행위를 논의하는 게시글을 수집하기가 어렵고, 다크웹 내에서의 언어 사용 특성에 대한 이해 부족으로 그동안 다크웹 사용자들이 어떤 내용의 범죄를 모의하는지 파악하기가 어려웠다. 본 논문에서는 국내 사용자들이 활동하는 다크웹 포럼들을 중심으로 사용자들의 언어 사용 특성을 연구하고, 이를 통해 다크웹에서 다뤄지는 범죄 유형들을 분석한다. 이를 위해, 자연어처리 기반의 분석 방법론을 적용하여 다크웹에서 공유되는 게시글을 수집하고 다크웹 사용자들의 은어와 특정 범죄군에서 선호되는 언어 특성을 파악한다. 특히 현재 다크웹 내에서 사용자들 사이에 관측되는 어휘들에 대한 기술통계 분석과 유의어 관계 분석을 수행하였고, 실제 다크웹 내에서 사용자들이 어떠한 범죄에 관심이 많은지를 분석하였으며, 더 나아가 수사의 효율성을 증대시키기 위한 소셜미디어, URL 인용 빈도에 대한 연구를 진행하였다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.7
• /
• pp.417-422
• /
• 2017

One of the most frightening aspects of weapons of mass destruction (WMD) is their ability to cause death in very small quantities without being visible to the public. The military authorities are making considerable effort to ensure the survivability of the combatants in the event of NBC(Nuclear, Biological and Chemical) contamination. Therefore, in this study, modules were developed for the measurement of the positive pressure and for the detection of the chemicals used for the control of the various shut-off valves used in an NBC shelter. In addition, a high performance gas tight shut-off valve was developed that can overcome the disadvantages associated with manual manufacturing, such as the occurrence of defective products and high manufacturing cost. By applying the positive pressure measurement and chemical detection modules, this valve was able to be used to control the facility. The developed gas-tight shut-off valve maintained airtight characteristics at a pressure loss of 28[Pa] at the prescribed wind velocity and an internal pressure of 30[kPa]. It is expected to be possible to control the gas-tight shut-off valve through the remote measurement of the positive pressure, thereby ensuring the foreign independence of import substitution and defense related technology in the future. In addition, by installing these valves in all of the intake ports or exhaust ports connected to the outside of the NBC shelter, it is possible to prevent the damage resulting from the rapid inflow of the storm pressure caused by conventional weapons and nuclear explosions, thereby protecting the people and equipment in the shelter.

• Journal of Astronomy and Space Sciences
• /
• v.22 no.4
• /
• pp.385-392
• /
• 2005

We developed an algorithm to efficiently detect streaks in survey images and made a performance test with YSTAR-NEOPAT images obtained by the 0.5m telescope stationed in South Africa. Fast moving objects whose apparent speeds exceed 10 arcsec/min are the main target of our algorithm; these include artificial satellites, space debris, and very fast Near-Earth Objects. Our algorithm, based on the outline shape of elongated sources employs a step of image subtraction in order to reduce the confusion caused by dense distribution of faint stars. It takes less than a second to find and characterize streaks present in normal astronomical images of 2K format. Comparison with visual inspection proves the efficiency and completeness of our automated detection algorithm. When applied to about 7,000 time-series images from YSTAR telescope, nearly 700 incidents of streaks are detected. Fast moving objects are identified by the presence of matching streaks in adjoining frames. Nearly all of confirmed fast moving objects turn out to be artificial satellites or space debris. Majority of streaks are however meteors and cosmic ray hits, whose identity is often difficult to classify.