• Journal of electromagnetic engineering and science
• /
• 제4권1호
• /
• pp.23-29
• /
• 2004

In this paper we analyze convergence constraints and time constant of IT-LMS algorithm and derive a method of making it's time constant independent of signal power by using input variance estimation. The method for estimating the input variance is to use a single-pole low-pass filter(LPF) with common smoothing parameter value, θ. The estimator is with narrow bandwidth for large θ but with wide bandwidth for small θ. This small θ gives long term average estimation(low frequency) of the fluctuating input variance well as short term variations (high frequency) of the input power. In our simulations of multipath communication channel equalization environments, the method with large θ has shown not as much improved convergence speed as the speed of the original IT-LMS algorithm. The proposed method with small θ=0.01 reach its minimum MSE in 100 samples whereas the IT-LMS converges in 200 samples. This shows the proposed, tap-power normalized IT-LMS algorithm can be applied more effectively to digital wireless communication systems.

• 천문학회보
• /
• 제44권1호
• /
• pp.54.2-54.2
• /
• 2019

Removing noise which occurs inevitably when taking image data has been a big concern. There is a way to raise signal-to-noise ratio and it is regarded as the only way, image stacking. Image stacking is averaging or just adding all pixel values of multiple pictures taken of a specific area. Its performance and reliability are unquestioned, but its weaknesses are also evident. Object with fast proper motion can be vanished, and most of all, it takes too long time. So if we can handle single shot image well and achieve similar performance, we can overcome those weaknesses. Recent developments in deep learning have enabled things that were not possible with former algorithm-based programming. One of the things is generating data with more information from data with less information. As a part of that, we reproduced stacked image from single shot image using a kind of deep learning, conditional generative adversarial network (cGAN). r-band camcol2 south data were used from SDSS Stripe 82 data. From all fields, image data which is stacked with only 22 individual images and, as a pair of stacked image, single pass data which were included in all stacked image were used. All used fields are cut in $128{\times}128$ pixel size, so total number of image is 17930. 14234 pairs of all images were used for training cGAN and 3696 pairs were used for verify the result. As a result, RMS error of pixel values between generated data from the best condition and target data were $7.67{\times}10^{-4}$ compared to original input data, $1.24{\times}10^{-3}$. We also applied to a few test galaxy images and generated images were similar to stacked images qualitatively compared to other de-noising methods. In addition, with photometry, The number count of stacked-cGAN matched sources is larger than that of single pass-stacked one, especially for fainter objects. Also, magnitude completeness became better in fainter objects. With this work, it is possible to observe reliably 1 magnitude fainter object.

• 한국통신학회논문지
• /
• 제24권8B호
• /
• pp.1518-1531
• /
• 1999

MPEG 및 H.263과 같은 동화상 압축 방식에 의하여 고 압축된 영상이 복원되었을 때 블록화 현상, 코너 이상치 (coner outliers), 링잉 잡음이 발생한다. 블록화 현상은 8x8 블록의 코너 (corner) 점에서 발생하는 잡음이며, 링잉 잡음은 영상의 에지 주변에서 발생하는 잡음이다. 그 이유는 MPEG 및 H.263이 8x8 화소 블록의 DCT 계수를 양자화 (quantization)하기 때문이다. 본 논문에서는 MPEG-4 및 H.263의 복원된 영상에서 발생하는 블록화 현상, 코니 이상치, 링잉 잡음을 줄이기 위해 기존의 저자들에 의하여 제안된 방법에 대한 주관적, 객관적 평가를 수행한다. 이 신호 적응형 후처리 방법은 압축된 데이터로부터 추출한 8x8 DCT 계수의 분포 정보와 움직임 벡터 정보를 이용하여 적응적으로 양자화 효과를 (quantization effect) 줄인다. 블록화 현상은 1차원 수평 및 수직 저대역 필터에 (low pass filter) 의하여 줄게 되고, 링잉 잡음은 2차원 신호 적응 필터 (signal-adaptive filter)에 의하여 줄게 된다. 신호 적응형 후처리 방법과 MPEG-4 VM (verification model)의 후처리 방법에 대한 MSSM (Modified Single Stimulus Method)을 이용한 주관적 \ulcorner질평가, 객관적 화질평가 (PSNR), 계산량 복잡도 (complexity)에 관한 비교연구가 컴퓨터 실험에 의하여 수행된다. 컴퓨터 실험을 위하여 MPEG-4에서 입력으로 사용하는 시험용 비데오 시퀀스를 이용하였다. 주관적인 화질 평가에서 두 방법은 비슷한 결과를 보였다. 반면 객관적 평가와 계산량 복잡도 분석 측면에 있어서, 신호 적응형 후처리 방법이 MPEG-4 VM의 후처리 방법보다 우수한 성능을 보였다.

• 한국CDE학회논문집
• /
• 제2권4호
• /
• pp.253-266
• /
• 1997

Pencil-curve machining, which is a single-pass ball-end milling along a concave edge on adie surface, is widely employed in die-surface machining. The cutter-path used for pencil-curve machining, which is the trajectory of the “ball-center point” of a ball-endmill sliding along a concave-edge region on the die surface, is called pencil-curve. Presented in the paper is a pencil-curve tracing algorithm in which “concave-type” sharp edges are computed from a “virtually digitized” model of the tool-envelope surface. The resulting “initial” pencil-cures are then refuted by applying a series of fairing operations. illustrative examples and methods for enhancing accuracy are also presented. The proposed pencil-curve tracing algorithm has been successfully implemented in a commercial CAM system specialized in die-machining and in the CAD/CAM system CATIA.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 Proceedings of the Korea Automatic Control Conference, 11th (KACC); Pohang, Korea; 24-26 Oct. 1996
• /
• pp.137-140
• /
• 1996

Finding line segments in an intensity image has been one of the most fundamental issues in computer vision. In complex scenes, it is hard to detect the locations of point features. Line features are more robust in providing greater positional accuracy. In this paper we present a robust "line features extraction" algorithm which extracts line feature in a single pass without using any assumptions and constraints. Our algorithm consists of five steps: (1) edge scanning, (2) edge normalization, (3) line-blob extraction, (4) line-feature computation, and (5) line linking. By using edge scanning, the computational complexity due to too many edge pixels is drastically reduced. Edge normalization improves the local quantization error induced from the gradient space partitioning and minimizes perturbations on edge orientation. We also analyze the effects of edge processing, and the least squares-based method and the principal axis-based method on the computation of line orientation. We show its efficiency with some real images.al images.

• 전기학회논문지P
• /
• 제67권1호
• /
• pp.42-46
• /
• 2018

In this paper, we propose fast object detection method of the cars by applying YOLOv2(You Only Look Once version 2) and UAVs (Unmanned Aerial Vehicles) while on the highway. We operated Darknet, OpenCV, CUDA and Deep Learning Server(SDX-4185) for our simulation environment. YOLOv2 is recently developed fast object detection algorithm that can detect various scale objects as fast speed. YOLOv2 convolution network algorithm allows to calculate probability by one pass evaluation and predicts location of each cars, because object detection process has simple single network. In our result, we could find cars on the highway area as fast speed and we could apply to the real time.

• 제어로봇시스템학회논문지
• /
• 제16권5호
• /
• pp.440-445
• /
• 2010

This paper proposes a single accelerometer sensor control algorithm to mobile inverted pendulum, generally called 'Segway', and evaluates the performance of this system comparing to the conventional ones. The commercialized 'Prototype Segway-PT' is initially considered as a next-generation transport vehicle. However, this robot is operated by three gyroscopes and two accelerometers to control the posture and speed, and it requires the complex signal processing for fusing the two sets of data. As the result of this, the growth rate of market size of 'Segway' is slow because of its high price mainly. In this paper, the mobile inverted pendulum is operated by a single accelerometer to simplify the control system to lower the price. Low pass filter is one of the good sensors to reducing the variation of an accelerometer, but it has time delay. This time delay disturbs real-time mobile inverted pendulum control. Like this, other various algorithms are used for this system, but each one has its own weak point. So this paper proposes a new filtering method, median filter and EKF. Median filter is used to image processing to reject impulse elements like salt and pepper noise. As the major performance evaluation parameter for the accelerometer, the high-frequency to low frequency ratio from FFT (Fast Fourier Transform) is used. Effectiveness of the proposed algorithms has been verified through the real experiments and the results are demonstrated.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2005년도 제36회 하계학술대회 논문집 B
• /
• pp.1480-1482
• /
• 2005

The major causes of power quality deterioration are harmonic current through semiconductor switching device, due to use of nonlinear loads such as diodes rectifier or thyristor rectifiers. In response to this concerns, this paper presents a new control method of single-phase active power filter(APF) for the compensation of harmonic current components in nonlinear loads. In order to make the complex calculation to be possible, the single-phase system that has two phases was made by constructing a imaginary second-phase giving time delay to load currents. In the conventional method, a imaginary-phase lagged to the load current T/4(here T is the fundamental cycle) was made. But in this proposed method, the new signal, which has the delayed phase through the filter, using the phase-delay property of low-pass filter, was used as the second phase. As this control method is applied to the system of single phase, an instantaneous calculation was done rather by using the rotating reference frames that synchronizes with source-frequency than by applying instantaneous reactive power theory that uses the conventional fixed reference frames.

• 한국HCI학회논문지
• /
• 제1권1호
• /
• pp.45-52
• /
• 2006

본 논문은 GPU를 활용한 이미지 공간 실시간 충돌 검사 기법을 설명한다. 닫힌 물체들이 충돌하지 않는 경우, 뷰잉 레이를 따라 물체의 앞면과 뒷면이 번갈아 가며 나타나는 것을 확인 할 수 있다. 그러나 물체 간 충돌이 일어나는 경우 이 현상이 깨어지게 된다. 이러한 특성에 기반하여 본 논문은 충돌 검사에 필요한 최소한의 표면 정보만 텍스쳐에 기록하여 충돌 검사를 수행하는 기법을 제안한다. 이 기법은 GPU의 framebuffer object 와 vertex buffer object, 그리고 occlusion query 등의 기능을 활용한다. 이러한 GPU의 기능을 이용하면 통상적인 이미지 기반 충돌검사에서 사용하는 multi-pass rendering 과 context switch 부하를 줄일 수 있다. 즉 기존의 이미지 기반 충돌 검사에 비해 적은 렌더링 횟수와 적은 렌더링 부하를 가진다. 본 논문에서 제안된 알고리즘은 변형체나 복잡한 물체에도 적용이 가능하며, 3D 게임이나 가상현실과 같은 실시간 어플리케이션에 적용될 수 있는 성능을 발휘한다.

• 전력전자학회논문지
• /
• 제20권2호
• /
• pp.182-192
• /
• 2015

This study develops a digital control scheme with power factor correction for a front-end converter in an electric vehicle battery charger. The front-end converter acts as the boost-type switching-mode rectifier. The converter assumes the two roles of the battery charger, which include power factor control and robust charging performance. The proposed control scheme consists of a charging control algorithm and a grid current control algorithm. The scheme aims to obtain unity input power factor and robust performance. Based on the linear average model of the converter, a constant-current constant-voltage charging control algorithm that passes through only one proportional-integral controller and a current feed-forward path is proposed. In the current control algorithm, we utilized a second band pass filter, a single-phase phase-locked loop technique, and a duty-ratio feed-forward term to control the grid current to be in phase with the grid voltage and achieve pure sinusoidal waveform. Simulations and experiments were conducted to verify the effectiveness of the proposed control scheme, both simulations and experiments.