• 한국HCI학회:학술대회논문집
• /
• 2008.02c
• /
• pp.172-177
• /
• 2008

Biometrics is essential for person identification because of its uniqueness from each individuals. Face recognition technology has advantage over other biometrics because of its convenience and non-intrusive characteristics. In this paper, we will present a overview of face recognition technology including face detection, feature extraction, and face recognition system. For face detection, we will describe template based method and face component based approach. PCA and LDA approach will be discussed for feature extraction, and nearest neighbor classifiers -will be covered for matching. Large database and the standardized performance evaluation methodology is essential in order to support state-of-the-art face recognition system. Also, 3D based face recognition technology is the key solution for the pose, lighting and expression variations in many applications.

• Proceedings of the KIEE Conference
• /
• 2003.07d
• /
• pp.2471-2473
• /
• 2003

인간의 안구는 색과 모양을 식별하는 데에 관여하는 원추 세포와 물체의 명암을 구분하는 간상 세포로 구성되어지는 시세포를 가진다. 망막 위에 분포되어 있는 시세포들은 시축을 중심으로 각기 다른 밀도로 분포 되어 있다. 특히 광축과 만나는 중심 지역은 Fovea라고 하는 직경 1mm 정도의 작은 우물을 이루고 있는데 원추 세포들이 고해상도로 분포되어 있고 시신경과 일대일로 연결되어 있어 시각 처리의 중심이 된다. 특히, 글자나 물체를 인식하기 위해 인간은 대상물을 응시하여 대상물의 영상이 Fovca영역에 맺히도록 추적 운동을 계속한다. 본 논문에서는 인간의 눈과 유사한 망막 모델의 하나인 Log-polar Image를 이용한 물체 인식을 위해 물체를 Log-polar Image Plane의 중심에 위치시키기 위한 모멘텀(Momentum)기반 정합 알고리즘(Alignment Algorithm)을 제시한다. 이는 눈동자 운동이 가능한 능동형 시각 장치의 Tracking 및 Pursuit 동작 중에 밭생하는 추적 오차를 보상함으로써 운동 중에도 효과적인 물체 인식이 가능하게 한다. 또한, 물체를 Log-polar Image Plane의 중심에 위치시킴으로써 물체의 위치 이동, 회전이동 그리고 크기 변화에 무관하게 물체를 인식한 수 있음을 제시한다.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.22 no.6
• /
• pp.735-744
• /
• 2019

One of the key factors for recognition performance in the automatic target recognition for synthetic aperture radar imagery(SAR-ATR) system is reliability of the SAR target database. To achieve optimal performance, the database should be constructed using the images obtained under the same operating condition as the SAR sensor. However, it is impractical to have the extensive set of real-world SAR images, and thus those from the electro magnetic prediction tool with 3-D CAD models are suggested as an alternative where their reliability can be always questionable. In this paper, a method for similarity assessment between target SAR images is presented inspired by the fact that a target SAR image is mainly characterized by the features of scattering centers. The method is demonstrated using a variety of examples and quantitatively measures the similarity related to reliability. Its assessment performance is further compared with that of the existing metric, structural similarity(SSIM).

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04b
• /
• pp.688-690
• /
• 2004

최근 전자화된 문서 영상을 효율적으로 관리하고 검색하기 위한 문서구조분석 방법과 문서의 자동 분류에 관한 많은 연구가 발표되고 있다. 본 논문에서는 NMF(non-negative matrix factorization) 알고리즘을 사용하여 폰트를 자동으로 분류하는 방법을 제안한다. 제안된 방법은 폰트의 구분 특징들이 공간적으로 국부성을 가지는 부분으로 표현될 수 있다는 가정을 바탕으로, 전체의 폰트 이미지들로부터 각 폰트들의 구분 특징인 부분을 학습하고, 학습된 부분들을 특징으로 사용하여 폰트를 분류하는 방법이다. 학습된 폰트의 특징들은 계층적 군집화 알고리즘을 이용하여 템플릿을 생성하고, 테스트 패턴을 분류하기 위하여 템플릿 패턴과의 EMD(earth mover's distance)를 사용한다. 실험결과에서 폰트 이미지들의 공간적으로 국부적인 특징들이 조사되고, 그 특징들의 폰트 식별을 위한 적절성을 보였다. 제안된 방법이 기존의 문자인식. 문서 검색 시스템들의 전처리기로 사용되면. 그 시스템들의 성능을 향상시킬 것으로 기대된다.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.45 no.2
• /
• pp.12-19
• /
• 2022

Artificial intelligence is driving the Fourth Industrial Revolution and is in the spotlight as a general-purpose technology. As the data collection from the battlefield increases rapidly, the need to us artificial intelligence is increasing in the military, but it is still in its early stages. In order to identify maritime targets, Republic of Korea navy acquires images by ISAR(Inverse Synthetic Aperture Radar) of maritime patrol aircraft, and humans make out them. The radar image is displayed by synthesizing signals reflected from the target after radiating radar waves. In addition, day/night and all-weather observations are possible. In this study, an artificial intelligence is used to identify maritime targets based on radar images. Data of radar images of 24 maritime targets in Republic of Korea and North Korea acquired by ISAR were pre-processed, and an artificial intelligence algorithm(ResNet-50) was applied. The accuracy of maritime targets identification showed about 99%. Out of the 81 warship types, 75 types took less than 5 seconds, and 6 types took 15 to 163 seconds.

• Journal of Korea Multimedia Society
• /
• v.24 no.12
• /
• pp.1624-1631
• /
• 2021

With the continued development of image processing technology, we live in a time when it is difficult to visually discriminate processed (or tampered) images from real images. However, as the risk of fake images being misused for crime increases, the importance of image forensic science for identifying fake images is emerging. Currently, various deep learning-based identifiers have been studied, but there are still many problems to be used in real situations. Due to the inherent characteristics of deep learning that strongly relies on given training data, it is very vulnerable to evaluating data that has never been viewed. Therefore, we try to find a way to improve generalization ability of deep learning-based fake image identifiers. First, images with various contents were added to the training dataset to resolve the over-fitting problem that the identifier can only classify real and fake images with specific contents but fails for those with other contents. Next, color spaces other than RGB were exploited. That is, fake image identification was attempted on color spaces not considered when creating fake images, such as HSV and YCbCr. Finally, dropout, which is commonly used for generalization of neural networks, was used. Through experimental results, it has been confirmed that the color space conversion to HSV is the best solution and its combination with the approach of increasing the training dataset significantly can greatly improve the accuracy and generalization ability of deep learning-based identifiers in identifying fake images that have never been seen before.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.34 no.3
• /
• pp.493-503
• /
• 2024

This paper addresses the issue of continuous user authentication within the metaverse environment. Recently, the metaverse plays a vital role in personal interaction, entertainment, education, and business, bringing forth significant security concerns. Particularly, vulnerabilities related to user identity verification have emerged as a major issue. This research proposes a novel method to verify identities by analyzing users' character movements in the metaverse through a pose estimation model. This method uses only video data for authentication, allowing flexibility in limited environments, and investigates how character movements contribute to user identification through various experiments. Furthermore, it explores the potential for extending this approach to other digital platforms. This research is expected to significantly contribute to enhancing security and innovating user identity verification methods in the metaverse environment.

• The Korean Journal of Nuclear Medicine
• /
• v.32 no.6
• /
• pp.490-496
• /
• 1998

Purpose: We evaluated brain perfusion SPECT findings of MELAS syndrome and mitochondrial myopathy in correlation with MR imaging in search of specific imaging features. Materials and Methods: Subjects were five patients (four females and one male; age range, 1 to 25 year) who presented with repeated stroke-like episodes, seizures or developmental delay or asymptomatic but had elevated lactic acid in CSF and serum. Conventional non-contrast MR imaging and Tc-99m-ethyl cysteinate dimer (ECD) brain perfusion SPECT were Performed and imaging features were analyzed. Results: MRI demonstrated increased T2 signal intensities in the affected areas of gray and white matters mainly in the parietal (4/5) and occipital lobes (4/5) and in the basal ganglia (1/5), which were not restricted to a specific vascular territory. SPECT demonstrated decreased perfusion in the corresponding regions of MRI lesions. In addition, there were perfusion defects in parietal (1 patient), temporal (2), and frontal (1) lobes and basal ganglia (1) and thalami (2). In a patient with mitochondrial myopathy who had normal MRI, decreased perfusion was noted in left parietal area and bilateral thalami. Conclusion: Tc-99m ECD SPECT imaging in patients with MELAS syndrome and mitochondrial myopathy showed hypoperfusion of parieto-occipital cortex, basal ganglia, thalamus and temporal cortex, which were not restricted to a specific vascular territory. There were no specific imaging features on SPECT. The significance of abnormal perfusion on SPECT without corresponding MR abnormalities needs to be evaluated further in larger number of patients.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.28 no.2
• /
• pp.311-326
• /
• 2018

This paper has experimented using News sample video with QCIF ($176{\times}144$) resolution in JM v10.2 code of H.264/AVC-MPEG. The region of interest (ROI) to be encrypted occurred the drift by unnecessarily referring to each frame continuously in accordance with the characteristics of the motion prediction and compensation of the H.264 standard. In order to mitigate the drift, the latest related research method of re-inserting encrypted I-picture into a certain period leads to an increase in the amount of additional computation that becomes the factor increasing the bit-rate overhead of the entire video. Therefore, the reference search range of the block and the frame in the ROI to be encrypted is restricted in the motion prediction and compensation for each frame, and the reference search range in the non-ROI not to be encrypted is not restricted to maintain the normal encoding efficiency. In this way, after encoding the video with restricted reference search range, this article proposes a method of RC4 bit-stream encryption for the ROI such as the face to be able to identify in order to protect personal information in the video. Also, it is compared and analyzed the experimental results after implementing the unencrypted original video, the latest related research method, and the proposed method in the condition of the same environment. In contrast to the latest related research method, the bit-rate overhead of the proposed method is 2.35% higher than that of the original video and 14.93% lower than that of the latest related method, while mitigating temporal drift through the proposed method. These improved results have verified by experiments of this study.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.12 no.3
• /
• pp.229-234
• /
• 2008

Hepatobiliary scintigraphy is very sensitivity of hepatic cell and gallbladder, biliary track atresia and biliary leakage. however, Hepatobiliary scan of biliary leakage diagnosis was separated determine biliary leakage and bowl drainage bile-juice. The object of this study will determine biliary leakage and bowl drainage bile-juice to hepatobiliary scintigraphy both decubitus position in bile leakage patients. Material & Methode: 31 patients (meal 14, Femeal 17), $51.1{\pm}14.4$ years. dynamic scan acquisition 60 farme for 60 minute on supine position. and delay scan was 2 hrs, 4 hrs, 24 hrs for 5 minute on supine, both decubitus position. Both decubitus position scan was kept for 5 minutes. Efficient of Hepatobiliary Scintigraphy both decubitus position in bile leakage patients was compared leakage size, density, image of supine position and both decubitus position. Results: 23 patients for 31 bile leakage patients was checked up function image or delay image, and 8 patients was checked up bile leakage on both decubitus. anatomical leakage location was supine position very well, but both decubitus position was separated bile leakage and moving bile-juice in bowl. also, uptake (counts/pixel) average of roi and bkg was supine 5.02, left decubitus 2.08, right decubitus 2.68. No. pixels of supine ROI counted 1.91 times than left decubitus, 1.05 times than right decubitus. Conclusion: 31 patient both decubitus position, but decubitus position was separated bile juice movement in bowl leakage location. also, It was compared ROI/BKG ratio and ROI No. pixels of supine, both decubitus in 38.5% patients. And No. pixels of supine position was large 19%, 5% than left decubitus, right decubitus, And density was in low 60%, 50% than left decubitus, right decubitus. It was mean bile leakage of ROI. so, If Hepatobiliary Scintigraphy was additional both decubitus position scan in bile leakage patients, this study will be more valuable in diagnosis of bile leakage.