• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.54 no.3
• /
• pp.166-173
• /
• 2005

Preserving image features, edges and details in the process of impulsive noise filtering is an important problem. To avoid image blurring, only corrupted pixels must be filtered. In this paper, we propose an effective impulse noise detection method using Self-Organizing Neural Network(SONN) which applies median filter selectively for removing random-valued impulse noises while preserving image features, edges and details. Using a $3\times3$ window, we obtain useful local features with which impulse noise patterns are classified. SONN is trained with sample image patterns and each pixel pattern is classified by its local information in the image. The results of the experiments with various images which are the noise range of $5-15\%$ show that our method performs better than other methods which use multiple threshold values for impulse noise detection.

• Proceedings of the KIEE Conference
• /
• 1998.07b
• /
• pp.410-413
• /
• 1998

A passive approach for enhancing fault detection and isolation performance of multiple observer based fault detection isolation schemes(FDIS) is proposed. The FDIS has a hierarchical framework to perform detection and isolation of faults of interest, and diagnosis of process faults. The decision unit comprises of a rule base and fuzzy inference engine and removes some difficulties of conventional decision unit which includes crisp logic and threshold values. Emphasis is placed on the design and evaluation methods of the diagnostic rule base. The suggested scheme is applied for the FDIS design for a DC motor driven centrifugal pump system.

• ETRI Journal
• /
• v.26 no.2
• /
• pp.92-100
• /
• 2004

In this paper, we propose an efficient method to broadcast digital television signals using single frequency networks (SFNs) within the Advanced Television Systems Committee (ATSC) transmission systems. In implementing the SFNs of an 8-vestigial side band (8-VSB) Digital Television (DTV) system, the ambiguity problem of the trellis coder is unavoidable in a conventional ATSC transmission system. We propose a memory initialization of the trellis coder to resolve this ambiguity problem. Since the proposed scheme to synchronize multiple transmitters minimizes the changes from the conventional ATSC system, the hardware complexity for these changes is very low. Our simulation results show that the proposed scheme makes a less than 0.1 dB degradation at the threshold of visibility with a bit error rate of $3{\times}10^{-6}$ in the additive white Gaussian noise (AWGN) channel. It is possible to reduce the performance degradation by increasing the initialization period of the proposed scheme.

• Biomaterials and Biomechanics in Bioengineering
• /
• v.3 no.1
• /
• pp.47-57
• /
• 2016

Results from multiple high profile experiments on the parameters influencing the impacts that cause skull fractures to the frontal, temporal, and parietal bones were gathered and analyzed. The location of the impact as a binary function of frontal or lateral strike, the velocity, the striking area of the impactor, and the force needed to cause skull fracture in each experiment were subjected to statistical analysis using the JMP statistical software pack. A novel neural network model predicting skull fracture threshold was developed with a high statistical correlation ($R^2=0.978$) and presented in this text. Despite variation within individual studies, the equation herein proposes a 3 kN greater resistance to fracture for the frontal bone when compared to the temporoparietal bones. Additionally, impacts with low velocities (<4.1 m/s) were more prone to cause fracture in the lateral regions of the skull when compared to similar velocity frontal impacts. Conversely, higher velocity impacts (>4.1 m/s) showed a greater frontal sensitivity.

• Journal of information and communication convergence engineering
• /
• v.11 no.3
• /
• pp.199-206
• /
• 2013

An emergency monitoring system for the elderly, which uses acceleration data measured with an accelerometer, angular velocity data measured with a gyroscope, and heart rate measured with an electrocardiogram, is proposed. The proposed fall detection algorithm uses multiple parameter combinations in which all parameters, calculated using tri-axial accelerations and bi-axial angular velocities, are above a certain threshold within a time period. Further, we propose an emergency detection algorithm that monitors the movements of the fallen elderly person, after a fall is detected. The results show that the proposed algorithms can distinguish various types of falls from activities of daily living with 100% sensitivity and 98.75% specificity. In addition, when falls are detected, the emergency detection rate is 100%. This suggests that the presented fall and emergency detection method provides an effective automatic fall detection and emergency alarm system. The proposed algorithms are simple enough to be implemented into an embedded system such as 8051-based microcontroller with 128 kbyte ROM.

• Journal of the Optical Society of Korea
• /
• v.8 no.4
• /
• pp.156-162
• /
• 2004

The spectroscopic model is proposed to analyze the gain bandwidth of a fiber Raman amplifier (FRA) with a multiple wavelength pumping scheme based on Raman gain theory. The oscillatory lineshape, which is the analytic function to analyze Raman gain spectra, allows us to estimate the gain bandwidth of the FRA. Based on the proposed theoretical modeling, we design and analyze the characteristics of the FRA using the combined multiwavelength pumping sources. We achieved the extended gain bandwidth of the FRA over 80 nm with the small gain ripple less than 0.5 dB. Threshold pumping power and effective noise figure for the FRA can be also analyzed by using the proposed model, which is also applicable for versatile fibers with other doping materials. The proposed analysis method can be useful for the design of FRA with the multiwavelength pumping scheme.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2013.08a
• /
• pp.243.2-243.2
• /
• 2013

본 실험에서 $SiO_2$, $SiN_x$ 게이트 절연막에 따른 ITZO 산화물 반도체 트랜지스터를 제작하여, 온도변화에 따라 전달 특성 변화를 측정하여 열에 대한 소자의 안정성을 비교, 분석하였다. 온도가 증가함에 따라 carrier가 증가하는 온도 의존성을 보이며, 이로 인해 Ioff가 증가하였다. multiple-trapping 모델을 적용하여, 이동도 증가와 문턱 전압이 감소를 확인하였다. 또한 M-N rule을 적용하여 $SiO_2$, $SiN_x$ 게이트 절연막을 가진 ITZO 산화물 박막 트랜지스터의 활성화 에너지를 추출하고, sub-threshold 지역에서 활성화 에너지의 변화량이 $SiO_2$, SiNX 각각 0.37 eV/V, 0.24 eV/V로 차이를 통해 $SiN_x$ 게이트 절연체를 가진 ITZO 산화물 반도체 트랜지스터의 이동도와 문턱 전압의 변화가 더 컸음을 확인하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.10b
• /
• pp.451-456
• /
• 2007

칩의 온도 상승에 대한 우려는 최근 점점 가시화되고 있다. 즉, 설계 집적도의 증가에 따른 전력 소모 밀도의 증가는 바로 칩 온도 상승으로 이어지고 있다. 이러한 칩 온도 상승은 성능 저하와 패키징 비용 증가 뿐 만 아니라, 칩의 신뢰성 칩 수명에서도 나쁜 악영향을 초래한다. 본 연구는 칩 온도 상승을 억제하기 위한 상위 단계 합성을 제안하고 있다. 구체적으로 본 연구의 핵심은 다중 전압 할당과 연산에 대한 모듈 바인딩(배정)을 동시에 고려한 새로운 저온도 설계 기법을 시도한다. 과거의 이중 threshold 전압 할당과 모듈 바인딩은 각각 누설 전류와 동적 전류를 줄이기 위해 적용된 반면 본 연구는 온도 최소화 측면에서 연구를 시도한 점에서 다른 설계 가능성을 보여 준다고 하겠다.

• Annual Conference of KIPS
• /
• 2004.05a
• /
• pp.691-694
• /
• 2004

본 논문에서는 실시간 처리에 적합한 효율적인 동영상 객체 분할 알고리즘을 제시한다. 제안된 동영상 객체 분할 알고리즘은 임계치 적용과 지역적 워터쉐드 알고리즘을 복합적으로 적용하였다. 첫째로 임계치 분할방법을 사용하여 초기 객체 마스크를 구성하였고 이러한 초기객체 마스크는 현재영상에서의 영역분할을 위한 입력으로 들어가게 된다. 최종적으로 지역적인 워터쉐드 분할방법을 초기 객체영역의 불명확한 지역에서만 다시 수행하여 최종적인 객체영역을 획득하여 기존 방식에 비해 분할시간을 줄였으며 분할성능을 높였다. 본 논문에서는 잡음환경에서 객체를 추출하기위해 복합적인 분할방식에 초점을 두었다. 이러한 복합적인 분할방법을 사용함으로써 객체 마스크 추출성능의 향상과 수행시간절약을 가져올 수 있었다.

• Current Optics and Photonics
• /
• v.4 no.6
• /
• pp.530-539
• /
• 2020

Hyperspectral images feature a relatively narrow band and are easily disturbed by noise. Accurate estimation of the types and parameters of noise in hyperspectral images can provide prior knowledge for subsequent image processing. Existing hyperspectral-noise estimation methods often pay more attention to the use of spectral information while ignoring the spatial information of hyperspectral images. To evaluate the noise in hyperspectral images more accurately, we have proposed a spectral-spatial cooperative noise-evaluation method. First, the feature of spatial information was extracted by Gabor-filter and K-means algorithms. Then, texture edges were extracted by the Otsu threshold algorithm, and homogeneous image blocks were automatically separated. After that, signal and noise values for each pixel in homogeneous blocks were split with a multiple-linear-regression model. By experiments with both simulated and real hyperspectral images, the proposed method was demonstrated to be effective and accurate, and the composition of the hyperspectral image was verified.