• Korean Journal of Remote Sensing
• /
• v.32 no.5
• /
• pp.471-482
• /
• 2016

This paper presents an extraction method for water body which uses block-based image partitioning and extension of water body boundaries to improve the performance of supervised classification for water body extraction. The Mahalanobis distance image is created by computing the spectral information of Normalized Difference Water Index (NDWI) and Near Infrared (NIR) band images over a training site within the water body in order to extract an initial water body area. To reduce the effect of noise contained in the Mahalanobis distance image, we apply mean curvature diffusion to the image, which controls diffusion coefficients based on connectivity strength between adjacent pixels and then extract the initial water body area. After partitioning the extracted water body image into the non-overlapping blocks of same size, we update the water body area using the information of water body belonging to water body boundaries. The update is performed repeatedly under the condition that the statistical distance between water body area belonging to water body boundaries and the training site is not greater than a threshold value. The accuracy assessment of the proposed algorithm was tested using KOMPSAT-2 images for the various block sizes between $11{\times}11$ and $19{\times}19$. The overall accuracy and Kappa coefficient of the algorithm varied from 99.47% to 99.53% and from 95.07% to 95.80%, respectively.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.6
• /
• pp.629-635
• /
• 2019

In Deep Learning models derivative is implemented by error back-propagation which enables the model to learn the error and update parameters. It can find the global (or local) optimal points of parameters even in the complex models taking advantage of a huge improvement in computing power. However, deliberately generated data points can 'fool' models and degrade the performance such as prediction accuracy. Not only these adversarial examples reduce the performance but also these examples are not easily detectable with human's eyes. In this work, we propose the method to detect adversarial datasets with random noise addition. We exploit the fact that when random noise is added, prediction accuracy of non-adversarial dataset remains almost unchanged, but that of adversarial dataset changes. We set attack methods (FGSM, Saliency Map) and noise level (0-19 with max pixel value 255) as independent variables and difference of prediction accuracy when noise was added as dependent variable in a simulation experiment. We have succeeded in extracting the threshold that separates non-adversarial and adversarial dataset. We detected the adversarial dataset using this threshold.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.1
• /
• pp.65-70
• /
• 2005

This paper deals with the problem of double-talk (DT) detection in anacoustic echo canceller (AEC). In the DT detection algorithm with correlation coefficient, detection errors occasionally occur because it is hard to set the threshold to distinguish DT from echo path change (EPC). Adaptive filter falls into the situation that it stops updating its filter coefficients when EPC is erroneously considered as DT at the starting-point of EPC. In addition, in case of echo path changing during the DT period, the end-point detection of DT period fails so that the AEC cannot update its filter coefficients for a while even after the DT period ends. To solve these problems, in this paper, we propose a novel AEC that employs an auxiliary filter. We use the idea that though the error signal cannot be estimated using reference signal in case or DT situation but it can be in case or EPC situation. The experimental result verifies that the proposed method could solve the problems caused by DT detection error or echo path change during the DT period.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.3
• /
• pp.77-82
• /
• 2018

This paper is concerned with the performance analysis of a modified stop-and-go generalized multi-modulus algorithm (MSAG-GMMA) adaptive blind equalization algorithm with variable step size. The proposed algorithm multiplies the fixed step size by the error signal of the decision-oriented algorithm in the equalization coefficient update equation, and changes the step size according to the error size. Also, the MSAG-GMMA having a fixed step size is operated so as to maintain a fast convergence speed from a certain threshold to a steady state by determining the error signal size of the decision-directed algorithm, and when the MSAG-GMMA to work To evaluate the performance of the proposed algorithm, we use the ensemble ISI, ensemble-averaged MSE, and equalized constellation obtained from the output of the equalizer as the performance index. Simulation results show that the proposed algorithm has faster convergence speeds than MMA, GMMA, and MSAG-GMMA and has a small residual error in steady state.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.49 no.7
• /
• pp.32-38
• /
• 2012

In this paper, to eliminate intersymbol interference of the received signal due to multipath propagation, a parallel equalization algorithm using two error estimation functions is proposed. In the proposed algorithm, multilevel two-dimensional signals are considered as equivalent binary signals, then error signals are estimated using the sigmoid nonlinearity effective at the initial phase equalization and threshold nonlinearity with high steady-state performance. The two errors are scaled by a weight depending on the relative accuracy of the two error estimations, then two filters are updated differentially. As a result, the combined output of two filters was to be the optimum value, fast convergence at initial stage of equalization and low steady-state error level were achieved at the same time thanks to the combining effect of two operation modes smoothly. Usefulness of the proposed algorithm was verified and compared with the conventional method through computer simulations.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.1
• /
• pp.46-53
• /
• 2011

Nonlocal means denoising algorithm is one of the most widely used denoising algorithm. Because it performs well, and the theoretic idea is intuitive and simple. However the conventional nonlocal means algorithm has still some problems such as noise remaining in the denoised flat region and blurring artifacts in the denoised edge and pattern region. Thus many improved algorithms based on nonlocal means have been proposed. In this paper, we proposed new improved nonlocal means denoising algorithm by weight update through weights sorting and newly defined threshold. Updated weights can make weights more refined and definite, and denoising is possible without that artifacts. Experimental results including comparisons with conventional algorithms for various noise levels and test images show the proposed algorithm has a good performance in both visual and quantitative criteria.

• Journal of Korea Game Society
• /
• v.14 no.5
• /
• pp.35-42
• /
• 2014

In the market of First Person Shooter (FPS) games, Massively Multi-player Online FPS games (MMOFPS) like 'PlanetSide 2' have been popular recently. Dead reckoning has been widely used in order to mitigate the network traffic overload for the game server with hundreds or thousands of people. This paper proposes the efficient analytical method to calculate the tolerable threshold angle of moving direction, which is one of the most important factors for character status updating when dead reckoning is used in MMOFPS games. The experimental results with game testers shows that the proposed method minimizes the position error for character moving and provides natural direction updates of characters.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• v.1
• /
• pp.277-282
• /
• 2006

It is well known that Kalman filtering is an optimal real-time data fusion method for GPS/INS integration. However, it has some limitations in terms of stability, adaptability and observability. A Kalman filter can perform optimally only when its dynamic model is correctly defined and the noise statistics for the measurement and process are completely known. It is found that estimated Kalman filter states could be influenced by several factors, including vehicle dynamic variations, filter tuning results, and environment changes, etc., which are difficult to model. Neural networks can map input-output relationships without apriori knowledge about them; hence a proper designed neural network is capable of learning and extracting these complex relationships with enough training. This paper presents a GPS/INS integrated system that combines Kalman filtering and neural network algorithms to improve navigation solutions during GPS outages. An Extended Kalman filter estimates INS measurement errors, plus position, velocity and attitude errors etc. Kalman filter states, and gives precise navigation solutions while GPS signals are available. At the same time, a multi-layer neural network is trained to map the vehicle dynamics with corresponding Kalman filter states, at the same rate of measurement update. After the output of the neural network meets a similarity threshold, it can be used to correct INS measurements when no GPS measurements are available. Selecting suitable inputs and outputs of the neural network is critical for this hybrid method. Detailed analysis unveils that some Kalman filter states are highly correlated with vehicle dynamic variations. The filter states that heavily impact system navigation solutions are selected as the neural network outputs. The principle of this hybrid method and the neural network design are presented. Field test data are processed to evaluate the performance of the proposed method.

• Journal of the Institute of Convergence Signal Processing
• /
• v.12 no.1
• /
• pp.13-19
• /
• 2011

In this paper, a visual surveillance system using SOPC based NIOS II embedded processor and C2H compiler was implemented. In this system, the IP is constructed by C2H compiler for the output of the camera images, image processing, serial communication and network communication, then, it is implemented to effectively control each IP based on the SOPC and the NIOS II embedded processor. And, an algorithm which updates the background images for high speed and robust detection of the moving objects is proposed using the Adaptive Gaussian Mixture Model(AGMM). In results, it can detecte the moving objects(pedestrians and vehicles) under day-time and night-time. It is confirmed that the proposed AGMM algorithm has better performance than the Adaptive Threshold Method(ATM) and the Gaussian Mixture Model(GMM) from our experiments.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.46 no.8
• /
• pp.22-30
• /
• 2009

Due to high levels of integration and complexity, the Network-on-Chip (NoC) approach has emerged as a new design paradigm to overcome on-chip communication issues and data bandwidth limits in conventional SoC(System-on-Chip) design. In particular, exponentially growing of energy consumption caused by high frequency, synchronization and distributing a single global clock signal throughout the chip have become major design bottlenecks. To deal with these issues, a globally asynchronous, locally synchronous (GALS) design combined with low power techniques is considered. Such a design style fits nicely with the concept of voltage-frequency-islands (VFI) which has been recently introduced for achieving fine-grain system-level power management. In this paper, we propose an efficient design methodology that minimizes energy consumption by VFI partitioning on an NoC architecture as well as assigning supply and threshold voltage levels to each VFI. The proposed algorithm which find VFI and appropriate core (or processing element) supply voltage consists of traffic-aware core graph partitioning, communication contention delay-aware tile mapping, power variation-aware core dynamic voltage scaling (DVS), power efficient VFI merging and voltage update on the VFIs Simulation results show that average 10.3% improvement in energy consumption compared to other existing works.