• Journal of the Korea Institute of Military Science and Technology
• /
• v.11 no.1
• /
• pp.66-74
• /
• 2008

This paper presents a new small target detection method using scale invariant feature. Detecting small targets whose sizes are varying is very important to automatic target detection. Scale invariant feature using the Laplacian scale-space can detect different sizes of targets robustly compared to the conventional spatial filtering methods with fixed kernel size. Additionally, scale-reflected adaptive thresholding can reduce many false alarms. Experimental results with real IR images show the robustness of the proposed target detection in real world.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.12 no.4
• /
• 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.

• East Asian mathematical journal
• /
• v.17 no.1
• /
• pp.103-114
• /
• 2001

When the target manifold is certain Kenmotsu manifolds, we characterize invariant immersions, tangential anti-invariant immersions and normal anti-invariant immersions by the spectra of the Jacobi operator.

• Journal of Institute of Control, Robotics and Systems
• /
• v.8 no.12
• /
• pp.991-997
• /
• 2002

The concept of feasible & invariant region plays an important role to derive closed loop stability and achie adequate performance of constrained receding horizon predictive control. In this paper, we define a complex polyhedral feasible & invariant set for all stabilizable input-constrained linear systems by using a complex transform and propose a one-norm based receding horizon control scheme using these invariant sets. In order to get a larger stabilizable set, a convex hull of invariant sets which are defined for different state feedback gains is used as a target invariant set of the constrained receding horizon control. The proposed constrained receding horizon control scheme is formulated so that it can be solved via linear programming.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.6
• /
• pp.198-207
• /
• 1996

A PSRI (position, scale, and rotation invariant ) feature extraction and automatic target recognition system using a cooperative network and an MLP is proposed. We can extract position invarient features by obtaining the target center using the projection and the moment in preprocessing stage. The scale and rotation invariant features are extracted from the contour projection of the number of edge pixels on each of the concentric circles, which is input to the cooperative network. By extracting the representative PSRI features form the features and their differentiations using max-net and min-net, we can rdduce the number of input neurons of the MLP, and make the resulted automatic target recognition system less sensitive to input variances. Experiments are conduted on various complex images which are shifted, rotated, or scaled, and the results show that the proposed system is very efficient for PSRI feature extractions and automatic target recognitions.

• Communications of the Korean Mathematical Society
• /
• v.12 no.2
• /
• pp.373-382
• /
• 1997

When the target manifold is a Sasakian or cosympletic space form, we characterize invariant immersions, tangential anti-invariant immersions and normal anti-invariant immersions by the spectra of the Jacobi operator.

• Proceedings of the KSRS Conference
• /
• 2009.03a
• /
• pp.105-110
• /
• 2009

현재 NASA에서 제공되는 MODIS 지표반사율자료(MOD09)는 MODIS영상을 이용한 각종 주제자료들의 중요한 입력 자료로 사용되고 있으며, MODIS 지표반사율 자료에 대한 객관적인 검증연구가 필요한 실정이다. 따라서 본 연구에서는 MOD09의 검증관련 초기 연구로서, 남한에 분포하는 불변성 타겟(invariant target)을 대상으로 2006년 일별 250m MODIS 지표반사율자료(MOD09GQK)자료의 객관적 검증을 시도하였다. 우선, MOD09 QA(Quality Assurance)자료를 이용하여 구름의 영향을 받은 화소를 제거한 후, 수치지도와 토지피복도를 이용하여 정의한 불변성 타겟에 해당되는 MOD09영상의 화소값을 추출하였다. 이와 같이 추출된 시계열 MOD09GHK영상의 화소값에 1차 회귀분석을 적용하여 이상 반사율 값을 탐지하고, 그 원인을 분석하였다. 검증 결과 나지지역에 대해서 0.0186의 RMSE값이 나타났으며, 인공물의 경우 0.2891의 RMSE값을 보였다. 발생된 이상 화소를 살펴보면, 구름, 그림자, 눈에 영향에 의해 발생한 것도 있으며, 원인을 알 수 없는 이상 화소들도 분포하였다. 향후 연구에서는 한반도 전역의 MODIS 시계열 반사율영상을 대상으로 MODIS 대기보정알고리즘과 입력인자의 적합성을 판단하기 위한 연구를 진행할 예정이다.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.25 no.10
• /
• pp.1062-1068
• /
• 2014

In this paper, a novel approach estimating target's rotation velocity(RV) is proposed for inverse synthetic aperture radar(ISAR) cross-range scaling(CRS). Scale invariant feature transform(SIFT) is applied to two sequently generated ISAR images for extracting non-fluctuating scatterers. Considering the fact that the distance between target's rotation center(RC) and SIFT features is same, we can set a criterion for estimating RV. Then, the criterion is optimized through the proposed method based on particle swarm optimization(PSO) combined with exhaustive search method. Simulation results show that the proposed algorithm can precisely estimate RV of a scenario based maneuvering target without RC information. With the use of the estimated RV, ISAR image can be correctly re-scaled along the cross-range direction.

• Journal of information and communication convergence engineering
• /
• v.21 no.2
• /
• pp.139-144
• /
• 2023

Engineers prefer deep neural networks (DNNs) for solving computer vision problems. However, DNNs pose two major problems. First, neural networks require large amounts of well-labeled data for training. Second, the covariate shift problem is common in computer vision problems. Domain adaptation has been proposed to mitigate this problem. Recent work on adversarial-learning-based unsupervised domain adaptation (UDA) has explained transferability and enabled the model to learn robust features. Despite this advantage, current methods do not guarantee the distinguishability of the latent space unless they consider class-aware information of the target domain. Furthermore, source and target examples alone cannot efficiently extract domain-invariant features from the encoded spaces. To alleviate the problems of existing UDA methods, we propose the mixup regularization in adversarial discriminative domain adaptation (ADDA) method. We validated the effectiveness and generality of the proposed method by performing experiments under three adaptation scenarios: MNIST to USPS, SVHN to MNIST, and MNIST to MNIST-M.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.39 no.4
• /
• pp.370-380
• /
• 2002

This paper proposes an ATR (Automatic Target Recognition) algorithm for identifying non-occluded and occluded military vehicles in natural FLIR (Forward Looking InfraRed) images. After segmenting a target, a radial function is defined from the target boundary to extract global shape features. Also, to extract local shape features of upper region of a target, a distance function is defined from boundary points and a line between two extreme points. From two functions and target contour, four global and four local shape features are proposed. They are much more invariant to translation, rotation and scale transform than traditional feature sets. In the experiments, we show that the proposed feature set is superior to the traditional feature sets with respect to the similarity-transform invariance and recognition performance.