• Proceedings of the IEEK Conference
• /
• 2003.07e
• /
• pp.2024-2027
• /
• 2003

본 논문에서는 얼굴 인식 분야에서 사용되는 PCA/LDA 알고리즘을 대신하기 위해 DCT/LDA 알고리즘을 제안하였다. PCA/LDA를 이용한 얼굴 인식의 경우 PCA 를 이용하여 얼굴 영상을 적은 수의 특징 값으로 표현한 다음 LDA를 수행한다. 그러나 PCA는 트레이닝 과정의 계산량이 많고 트레이닝 셋이 변할 때마다 기저 벡터가 변화한다. PCA/LDA의 단점을 개선하기 위해 계산량이 적고 기저 벡터가 일정한 DCT의 계수를 사용한다. DCT/LDA를 사용할 경우 특징 값을 빠르게 추출하면서 PCP/LDA와 유사한 성능을 얻을 수 있다. 실험을 통하여 포즈 변화와 조명 변화가 있는 얼굴 데이터 셋에서 최고 97.8%의 인식률을 보였다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.5A
• /
• pp.717-725
• /
• 2000

Within the last several years, there has been a large number of algorithms developed for face recognition. The majority of these algorithms have been view- and projection-based algorithms. Our definition of projection is not restricted to projecting the image onto an orthogonal basis the definition is expansive and includes a general class of linear transformation of the image pixel values. The class includes correlation, principal component analysis, clustering, gray scale projection, and matching pursuit filters. In this paper, we perform a detailed analysis of this class of algorithms by evaluating them on the FERET database of facial images. In our experiments, a projection-based algorithms consists of three steps. The first step is done off-line and determines the new basis for the images. The bases is either set by the algorithm designer or is learned from a training set. The last two steps are on-line and perform the recognition. The second step projects an image onto the new basis and the third step recognizes a face in an with a nearest neighbor classifier. The classification is performed in the projection space. Most evaluation methods report algorithm performance on a single gallery. This does not fully capture algorithm performance. In our study, we construct set of independent galleries. This allows us to see how individual algorithm performance varies over different galleries. In addition, we report on the relative performance of the algorithms over the different galleries.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.14 no.2
• /
• pp.439-446
• /
• 2019

In this paper, we propose a encryption method using C-MLCA and 1D CAT to secure medical image for efficiently. First, we generate a state transition matrix using a Wolfram rule and create a sequence of maximum length. By operating the complemented vector, it converts an existing sequence to a more complex sequence. Then, we multiply the two sequences by rows and columns to generate C-MLCA basis images of the original image size and go through a XOR operation. Finally, we will get the encrypted image to operate the 1D CAT basis function created by setting the gateway values and the image which is calculated by transform coefficients. By comparing the encrypted image with the original image, we evaluate to analyze the histogram and PSNR. Also, by analyzing NPCR and key space, we confirmed that the proposed encryption method has a high level of stability and security.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.1
• /
• pp.214-223
• /
• 1996

Noise effects on image profiles reconstructed by the spectral inverse scattering technique is studied based on moment method with series-expanded basis function. It is found that the Fourier series expansion to the field distribution and the averaging of the reconstructed profile in each enlarged cell provides an effective tool for the reduction of noise effects.

• Journal of the Korean Society of Radiology
• /
• v.82 no.5
• /
• pp.1281-1286
• /
• 2021

Brainstem infarction due to vertebrobasilar insufficiency is a rare initial presentation of systemic lupus erythematosus (SLE) patients and small-vessel dissection as the direct cause of infarction has not been reported. We report the case of a 20-year-old female with acute infarction on the right side of the pons due to a small artery (pontine perforator) dissection, identified on digital subtraction angiography and high-resolution vessel wall MRI (vwMRI). She was diagnosed with SLE based on the presence of neurologic disorders and relevant laboratory findings. The pontine perforator-dissecting aneurysm had occluded and the right distal vertebral artery had resolved on subsequent vwMRI. She had a modified Rankin Scale (mRS) score of 1 at discharge with mild symptom improvement, and exhibited no further aggravation of symptoms at 3 or 12 months, maintaining an mRS score of 1.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.20 no.6
• /
• pp.568-573
• /
• 2000

• Proceedings of the KIEE Conference
• /
• 2005.10b
• /
• pp.349-351
• /
• 2005

현재 다양한 인터넷 콘텐츠들에 의해 많은 정보가 공유되고 있으며, 유익한 정보들과 더불어 성인물과 같은 유해한 정보들이 있다. 이로 인하여 여러 문제점들이 야기되고 있으며, 이를 해결하기 위해 다양한 방법들이 제안되고 있다. 그 중에서 성인 영상 차단을 위한 연구도 많이 행해지고 있으며 주로 색상을 이용한 방법을 사용하고 있다. 그러나 살색과 유사한 영상이나 노출이 심한 영상에는 성인 영상 검출의 신뢰성이 떨어지는 단점을 갖는다. 본 논문에서는 이런 문제점을 해결하기 위해 새로운 성인 영상 차단 방법을 제안한다. 기존의 제안된 살색 검출을 이용한 방법을 기반으로 성인 영상물로 판정될 수 있는 신체 부위를 검출함으로써 강인한 성인 영상 차단을 한다. 신체 부위에 대한 판별을 위해 여러 기저 영상에서 특징 벡터를 추출하고. 이 벡터를 Back Propagation(BP) 신경망의 데이터로 이용하여 학습한다. 제안한 성인 영상 차단 방법의 성능을 여러 장의 살색과 유사한 색상의 물체 영상과 노출이 심한 영상, 성인 영상을 이용한 종합적인 실험 결과인 성인 영상 검출률을 통해 증명한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.07a
• /
• pp.294-296
• /
• 2010

일반적으로, RGB 영상의 높은 주파수 영역은 잡음으로 인해 색평면 간 서로 낮은 상관도를 가지고 있기 때문에 이러한 고주파수 성분은 색평면 간 예측의 효율을 저하시키는 원인이 된다. 본 논문에서는 RGB 비디오 코딩에서 색평면 간 예측의 효율을 높이기 위해 기저 색평면을 적응적으로 필터링 하는 방법을 제안한다. 색평면 간 상관도에 따라 적응적으로 기저 색평면을 필터링함으로써 색평면 간 예측 성능을 높일 수 있었다. 본 논문에서 제안하는 알고리즘을 통해 우리는 H.264/AVC High 4:4:4 Intra Profile에 비해 평균 14.71%의 비트율 감소와 0.93dB의 PSNR 향상 결과를 얻을 수 있었다.

• Investigative Magnetic Resonance Imaging
• /
• v.4 no.2
• /
• pp.100-106
• /
• 2000

Purpose : To evaluate the usefulness of cerebral blood flow measurement applied to perfusion weighted image with short-scan time single shot gradient echo-planar technique in measuring cerebral blood volume(rCBV) of normal rabbits. Materials and methods : With 2.1-3.6 kg weighted rabbits, image is acquired when they are in supine position in children positioner. Perfusion weighted image is acquired to 44 seconds per 1 second successively. After 4 seconds later, Gd-DTPA 2ml are injected into int. jugular vein with 2 ml per second and normal saline is also injected after that. Same technique is applied 2 times per 30 minites in same rabbit. After Image is obtained in two part of cerebral cortex at vertex, convexity, in one of basal ganglia with choosing about $3-5{\textrm{mm}^2}$ areas. Curve of signal intensity changes in time sequence is drawn. After this images are transmitted by PC and software IDL, regional cerebral blood volume is measured with imaging processing program made by us. Results : With 22 of 24 rabbits, satisfactory 1-2 signal intensity versus time curve is made. Cerebral blood capacity and contrast media stay time (ST) is measured in two cerebral cortex and basal ganglia refering in parietal cerebral cortex. Mean focal cerebral blood flow capacity ratio in cortex was $0.97{\pm}0.35$ and in basal ganglia, $0.99{\pm}0.37$, mean contrast media stay time in cortex was $9.83{\pm}1.63$ sec and in basal gaiglia, $9.42{\pm}1.14$ sec, but there was no statistically significant difference between two areas ($\rho$=0.05). Conclusion : In cerebral cortex and basal ganglia, there is no difference in mean focal blood volume and mean contrast stay time. Therefore, PWI is useful in cerebral blood flow and early diagnosis, prognosis of cerebral ischemic disease. Hereafter, it is helpful in analysing cerebral blood flow changes with comparison difference in rCBV between normal tissue and ischemic tissue, and that with DWI finding in infarcted patient.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.6C
• /
• pp.361-367
• /
• 2011

In this paper, we propose encryption method to using complemented MLCA(Maximum Length Cellular Automata) based on NBCA(Null Boundary CA) and 2D CAT (Two-Dimensional Cellular Automata Transform) for efficient image encryption. The encryption method is processed in the following order. First, a transition matrix T is created using the Wolfram Rule matrix. Then, the transition matrix T is multiplied to the original image that is intended to be encrypted, which transfers the pixel values of the original image. Furthermore, the converted original image goes through a XOR operation with complemented vector F to convert into a complemented MLCA applied image. Then, the gateway value is set and 2D CAT basis function is created. Also, the 2D CAT is encrypted by multiplying the created basis function to the complemented MLCA applied image. Lastly, the stability analysis verifies that proposed method holds a high encryption quality status.