• 대한임베디드공학회논문지
• /
• 제12권4호
• /
• pp.231-238
• /
• 2017

Detecting small infrared targets from the low-SCR images at a long distance is very hard. The previous Local Contrast Method (LCM) algorithm based on the human visual system shows a superior performance of detecting small targets by a background suppression technique through local contrast measure. However, its slow processing speed due to the heavy multi-scale processing overhead is not suitable to a variety of real-time applications. This paper presents a lightweight real-time small target detection algorithm, called by the Improved Selective Local Contrast Method (ISLCM), to reduce the scale-invariant computational overhead. The proposed ISLCM applies the improved local contrast measure to the predicted selective region so that it may have a comparable detection performance as the previous LCM while guaranteeing low scale-invariant computational load by exploiting both adaptive scale estimation and small target feature feasibility. Experimental results show that the proposed algorithm can reduce its computational overhead considerably while maintaining its detection performance compared with the previous LCM.

• 대한수학회논문집
• /
• 제15권4호
• /
• pp.649-668
• /
• 2000

In this paper we prove the following : Let M be a real (2n-1)-dimensional compact minimal semi-invariant submanifold in a complex projective space P(sub)n+1C. If the scalar curvature $\geq$2(n-1)(2n+1), then m is a homogeneous type $A_1$ or $A_2$. Next suppose that the third fundamental form n satisfies dn = 2$\theta\omega$ for a certain scalar $\theta$$\neq$c/2 and $\theta$$\neq$c/4 (4n-1)/(2n-1), where $\omega$(X,Y) = g(X,øY) for any vectors X and Y on a semi-invariant submanifold of codimension 3 in a complex space form M(sub)n+1 (c). Then we prove that M has constant principal curvatures corresponding the shape operator in the direction of the distingusihed normal and the structure vector ξ is an eigenvector of A if and only if M is locally congruent to a homogeneous minimal real hypersurface of M(sub)n (c).

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제4권6호
• /
• pp.1253-1272
• /
• 2010

In this paper, we propose a new affine shape adaptation scheme for the affine invariant feature detector, in which the convergence stability is still an opening problem. This paper examines the relation between the integration scale matrix of next iteration and the current second moment matrix and finds that the convergence stability of the method can be improved by adjusting the relation between the two matrices instead of keeping them always proportional as proposed by previous methods. By estimating and updating the shape of the integration kernel and differentiation kernel in each iteration based on the anisotropy of the current second moment matrix, we propose a coarse-to-fine affine shape adaptation scheme which is able to adjust the pace of convergence and enable the process to converge smoothly. The feature matching experiments demonstrate that the proposed approach obtains an improvement in convergence ratio and repeatability compared with the current schemes with relatively fixed integration kernel.

• 한국지능시스템학회논문지
• /
• 제23권2호
• /
• pp.178-183
• /
• 2013

In this study, we proposed a method for electroencephalogram (EEG) classification using invariant CSP at special channels for improving the accuracy of classification. Based on the naive EEG signals from left and right hand movement experiment, the noises of contaminated data set should be eliminate and the proposed method can deal with the de-noising of data set. The considering data set are collected from the special channels for right and left hand movements around the motor cortex area. The proposed method is based on the fit of the adjusted parameter to decline the affect of invariant parts in raw signals and can increase the classification accuracy. We have run the simulation for hundreds time for each parameter and get averaged value to get the last result for comparison. The experimental results show the accuracy is improved more than the original method, the highest result reach to 89.74%.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제4권3호
• /
• pp.322-326
• /
• 2004

This paper describes a new iris recognition method based on a shift-invariant wavelet sub-images. For the feature representation, we first preprocess an iris image for the compensation of the variation of the iris and for the easy implementation of the wavelet transform. Then, we decompose the preprocessed iris image into multiple subband images using a shift-invariant wavelet transform. For feature representation, we select a set of subband images, which have rich information for the classification of various iris patterns and robust to noises. In order to reduce the size of the feature vector, we quantize. each pixel of subband images using the Lloyd-Max quantization method Each feature element is represented by one of quantization levels, and a set of these feature element is the feature vector. When the quantization is very coarse, the quantized level does not have much information about the image pixel value. Therefore, we define a new similarity measure based on mutual information between two features. With this similarity measure, the size of the feature vector can be reduced without much degradation of performance. Experimentally, we show that the proposed method produced superb performance in iris recognition.

• 제어로봇시스템학회논문지
• /
• 제18권2호
• /
• pp.149-155
• /
• 2012

This paper provides an efficient model predictive control for the output voltage control of three-phase inverter system which includes output LC filters. Use of SVPWM (Space Vector Pulse-Width-Modulation) and the rotating d-q frame is made to obtain an input constrained dynamic model of the inverter system. From the measured/estimated output current and reference output voltage, corresponding equilibrium values of the inductor current and the control input are computed. Derivation of a feasible and invariant set around the equilibrium state is made and then a receding horizon strategy which steers the current state deep into the invariant set is proposed. In order to remove offset error, use of disturbance observer is made in the form of state estimator. The efficacy of the proposed method is verified through simulations.

• 대한임베디드공학회논문지
• /
• 제17권6호
• /
• pp.309-317
• /
• 2022

In this paper, we propose a machine learning-based copy-move forgery detection network with dual branches. Because the rotation or scaling operation is frequently involved in copy-move forger, the conventional convolutional neural network is not effectively applied in detecting copy-move tampering. Therefore, we divide the input into rotation-invariant and scaling-invariant features based on the wavelet coefficients. Each of the features is input to different branches having the same structure, and is fused in the combination module. Each branch comprises feature extraction, correlation, and mask decoder modules. In the proposed network, VGG16 is used for the feature extraction module. To check similarity of features generated by the feature extraction module, the conventional correlation module used. Finally, the mask decoder model is applied to develop a pixel-level localization map. We perform experiments on test dataset and compare the proposed method with state-of-the-art tampering localization methods. The results demonstrate that the proposed scheme outperforms the existing approaches.

• 한국컴퓨터정보학회논문지
• /
• 제28권8호
• /
• pp.59-66
• /
• 2023

골든크로스를 흔히 매수의 신호로 인식하지만, 주식시장은 변동성이 매우 크기에 골든크로스만으로 주식의 등락 여부를 예상하고 의사결정을 내리기에는 무리가 있다. 마찬가지로, 이러한 주가 데이터의 불확실성은 기존의 시계열 기반의 예측을 더욱 어렵게 한다. 본 논문에서는 골든크로스를 하나의 사건으로 인식하여, time-invariant 한 접근을 시도하고자 한다. LSTM 신경망 기법을 사용하여 골든크로스 이후의 주가 변화율을 예측하고, 기존의 시계열 분석에서 도출한 성능과 종목별로 비교한다. 또한, 0을 기준으로 한 주가 변화율의 등락을 혼동행렬로 분류하여 일반화 분류 성능을 입증한다. 최종적으로 본 논문은 예측 정밀도가 83%인 모델을 제안하였다. 골든크로스가 나타날 때 모든 상황에서 매수를 결정하기보다 모델을 활용하여 투자자의 투자 자본 손실을 방지할 수 있다.

• 대한수학회보
• /
• 제22권1호
• /
• pp.7-10
• /
• 1985

For each a.mem.S and f.mem.m(S), denote by $l_{a}$ f(s)=f(as) for all s.mem.S. If A is a norm closed left translation invariant subalgebra of m(S) (i.e. $l_{a}$ f.mem.A whenever f.mem.A and a.mem.S) containing 1, the constant ont function on S and .phi..mem. $A^{*}$, the dual of A, then .phi. is a mean on A if .phi.(f).geq.0 for f.geq.0 and .phi.(1) = 1, .phi. is multiplicative if .phi. (fg)=.phi.(f).phi.(g) for all f, g.mem.A; .phi. is left invariant if .phi.(1sf)=.phi.(f) for all s.mem.S and f.mem.A. It is well known that the set of multiplicative means on m(S) is precisely .betha.S, the Stone-Cech compactification of S[7]. A subalgebra of m(S) is (extremely) left amenable, denoted by (ELA)LA if it is nom closed, left translation invariant containing contants and has a multiplicative left invariant mean (LIM). A semigroup S is (ELA) LA, if m(S) is (ELA)LA. A subset E.contnd.S is left thick (T. Mitchell, [4]) if for any finite subser F.contnd.S, there exists s.mem.S such that $F_{s}$ .contnd.E or equivalently, the family { $s^{-1}$ E : s.mem.S} has finite intersection property.y.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제55권4호
• /
• pp.173-180
• /
• 2006

A key component of autonomous navigation of intelligent home robot is localization and map building with recognized features from the environment. To validate this, accurate measurement of relative location between robot and features is essential. In this paper, we proposed relative localization algorithm based on 3D reconstruction of scale invariant features of two images which are captured from two parallel cameras. We captured two images from parallel cameras which are attached in front of robot and detect scale invariant features in each image using SIFT(scale invariant feature transform). Then, we performed matching for the two image's feature points and got the relative location using 3D reconstruction for the matched points. Stereo camera needs high precision of two camera's extrinsic and matching pixels in two camera image. Because we used two cameras which are different from stereo camera and scale invariant feature point and it's easy to setup the extrinsic parameter. Furthermore, 3D reconstruction does not need any other sensor. And the results can be simultaneously used by obstacle avoidance, map building and localization. We set 20cm the distance between two camera and capture the 3frames per second. The experimental results show :t6cm maximum error in the range of less than 2m and ${\pm}15cm$ maximum error in the range of between 2m and 4m.