• Korean Journal of Remote Sensing
• /
• v.24 no.5
• /
• pp.417-425
• /
• 2008

An image-based atmospheric correction software ATC is implemented using MATLAB and SML (Spatial Modeler Language in ERDAS IMAGINE), and it was tested using Landsat TM/ETM+ data. This ATC has two main functional modules, which are composed of a semiautomatic type and an automatic type. The semi-automatic functional module includes the Julian day (JD), Earth-Sun distance (ESD), solar zenith angle (SZA) and path radiance (PR), which are programmed as individual small functions. For the automatic functional module, these parameters are computed by using the header file of Landsat TM/ETM+. Three atmospheric correction algorithms are included: The apparent reflectance model (AR), one-percent dark object subtraction technique (DOS), and cosine approximation model (COST). The ACT is efficient as well as easy to use in a system with MATLAB and SML.

• Communications for Statistical Applications and Methods
• /
• v.26 no.1
• /
• pp.69-78
• /
• 2019

A structured model with both single-index and varying coefficients is a powerful tool in modeling high dimensional data. It has been widely used because the single-index can overcome the curse of dimensionality and varying coefficients can allow nonlinear interaction effects in the model. For high dimensional index vectors, variable selection becomes an important question in the model building process. In this paper, we propose an efficient estimation and a variable selection method based on a smoothing spline approach in a partially linear single-index-coefficient regression model. We also propose an efficient algorithm for simultaneously estimating the coefficient functions in a data-adaptive lower-dimensional approximation space and selecting significant variables in the index with the adaptive LASSO penalty. The empirical performance of the proposed method is illustrated with simulated and real data examples.

• Journal of Broadcast Engineering
• /
• v.26 no.2
• /
• pp.125-131
• /
• 2021

In recent years, Convolutional Neural Networks (CNNs) have achieved outstanding performance in the fields of computer vision such as image classification, object detection, visual quality enhancement, etc. However, as huge amount of computation and memory are required in CNN models, there is a limitation in the application of CNN to low-power environments such as mobile or IoT devices. Therefore, the need for neural network compression to reduce the model size while keeping the task performance as much as possible has been emerging. In this paper, we propose a method to compress CNN models by combining matrix decomposition methods of LR (Low-Rank) approximation and CP (Canonical Polyadic) decomposition. Unlike conventional methods that apply one matrix decomposition method to CNN models, we selectively apply two decomposition methods depending on the layer types of CNN to enhance the compression performance. To evaluate the performance of the proposed method, we use the models for image classification such as VGG-16, RestNet50 and MobileNetV2 models. The experimental results show that the proposed method gives improved classification performance at the same range of 1.5 to 12.1 times compression ratio than the existing method that applies only the LR approximation.

• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.233-236
• /
• 2000

The acoustical response of fish depends on size and physical structure na, most important, on the presence or absence of a swimbladder. Acoustic scattering models for swimbladdered fish represent a fish by an ideal pressure-release surface having the size and shape as the swimbladder. Target strength experiments of red seabream (Chrysophrys major) have been conducted using 38 (split-beam), 120 (split-beam) and 200kHz (dual-beam) frequencies. At each start of each experiment, the live fish are placed in the cage at the surface, then the cage is lowed to about $4{\cal}m$ depth where it remains during the measurements. To test the acoustic models, predictions of target strength based on swimbladder morphometries of 10 red seabream offish total length from $103{\cal}mm{\;}to{\;}349{\cal}mm$ ($3 <$TL/\lambda$ < 45)are compared with conventional target strength measurements on the same, shock-frozen immediately after caged experiments. X-ray was projected along dorsal aspect to know the morphological construction of swimbladder. and fish body. At high frequencies, Helmholtz-kirchhoff(HK) approximation would greatly enhance swimbladdered fish modeling. Sound scattering model [HK-ray approximation model] for comparison to experimental target strength data was used to model backscatter measurements from individual fish. The scattering data can be used in the inverse method along with multiple frequency sonar systems to investigate the adequacy of classification and identification of fish

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.50 no.7
• /
• pp.339-349
• /
• 2001

This paper suggests an optimal identification method for complex and nonlinear system modeling that is based on Fuzzy-Neural Networks(FNN). The proposed Hybrid Identification Algorithm is based on Yamakawa's FNN and uses the simplified inference as fuzzy inference method and Error Back Propagation Algorithm as learning rule. In this paper, the FNN modeling implements parameter identification using HCM algorithm and hybrid structure combined with two types of optimization theories for nonlinear systems. We use a HCM(Hard C-Means) clustering algorithm to find initial apexes of membership function. The parameters such as apexes of membership functions, learning rates, and momentum coefficients are adjusted using hybrid algorithm. The proposed hybrid identification algorithm is carried out using both a genetic algorithm and the improved complex method. Also, an aggregated objective function(performance index) with weighting factor is introduced to achieve a sound balance between approximation and generalization abilities of the model. According to the selection and adjustment of a weighting factor of an aggregate objective function which depends on the number of data and a certain degree of nonlinearity(distribution of I/O data), we show that it is available and effective to design an optimal FNN model structure with mutual balance and dependency between approximation and generalization abilities. To evaluate the performance of the proposed model, we use the time series data for gas furnace, the data of sewage treatment process and traffic route choice process.

• The Korean Journal of Applied Statistics
• /
• v.27 no.3
• /
• pp.397-406
• /
• 2014

The minimum density power divergence estimation has been a popular topic in the field of robust estimation for since Basu et al. (1988). The minimum density power divergence estimator has strong robustness properties with the little loss in asymptotic efficiency relative to the maximum likelihood estimator under model conditions. However, a limitation in applying this estimation method is the algebraic difficulty on an integral involved in an estimation function. This paper considers a minimum density power divergence estimation method with approximated divergence avoiding such difficulty. As an example, we consider the normal-exponential convolution model introduced by Bolstad (2004). The estimated divergence in this case is too complicated; consequently, a Laplace approximation is employed to obtain a manageable form. Simulations and an empirical study show that the minimum density power divergence estimators based on an approximated estimated divergence for the normal-exponential model perform adequately in terms of bias and efficiency.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.10
• /
• pp.5015-5038
• /
• 2018

In this paper, a novel spectral-spatial joint sparse representation algorithm for hyperspectral image classification is proposed based on multi-layer superpixels in various scales. Superpixels of various scales can provide complete yet redundant correlated information of the class attribute for test pixels. Therefore, we design a joint sparse model for a test pixel by sampling similar pixels from its corresponding superpixels combinations. Firstly, multi-layer superpixels are extracted on the false color image of the HSI data by principal components analysis model. Secondly, a group of discriminative sampling pixels are exploited as reconstruction matrix of test pixel which can be jointly represented by the structured dictionary and recovered sparse coefficients. Thirdly, the orthogonal matching pursuit strategy is employed for estimating sparse vector for the test pixel. In each iteration, the approximation can be computed from the dictionary and corresponding sparse vector. Finally, the class label of test pixel can be directly determined with minimum reconstruction error between the reconstruction matrix and its approximation. The advantages of this algorithm lie in the development of complete neighborhood and homogeneous pixels to share a common sparsity pattern, and it is able to achieve more flexible joint sparse coding of spectral-spatial information. Experimental results on three real hyperspectral datasets show that the proposed joint sparse model can achieve better performance than a series of excellent sparse classification methods and superpixels-based classification methods.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.51 no.3
• /
• pp.405-413
• /
• 2015

Acoustic target strength (TS) measurement and theoretical acoustic scattering models were applied to estimate the TS for assessing the fish biomass. TS measurement was made of ex situ sailfin sandfish Arctoscopus japonicus at 70 and 120 kHz, and then compared to backscatter model prediction. The live sandfish was tethered in seawater using monofilament lines. Measurements were made versus incidence angle, $-50^{\circ}$ to $50^{\circ}$ relative to dorsal aspect directions. Distorted wave Born approximation (DWBA) model was used to calculated TS pattern. The TS values of sandfish (body length: 16.2 cm) at 70 and 120 kHz were -66.94 dB and -64.45 dB, respectively, and were about 20 dB lower than TS of other fishes in Korea waters. These TS levels were distributed within the range of the theoretical TS. Ex situ measurement and theoretical TS may be applied to improve acoustical detection and biomass estimation of the sandfish, and is necessary to measure with various lengths.

• Journal of KIISE:Software and Applications
• /
• v.26 no.10
• /
• pp.1143-1155
• /
• 1999

셀룰러 방식의 이동통신 시스템에서 전파의 유효신호 도달범위를 예측하기 위해서는 전파전파 모델을 이용한 예측기법이 주로 사용된다. 그러나, 전파과정에서 주변 지형지물에 의해 발생하는 전파손실은 매우 복잡한 비선형적인 특성을 가지며 수식으로는 정확한 표현이 불가능하다. 본 논문에서는 신경회로망의 함수 근사화 능력을 이용하여 전파손실 예측모델을 생성하는 방법을 제안한다. 즉, 전파손실을 송수신 안테나간의 거리, 송신안테나의 특성, 장애물 투과영향, 회절특성, 도로, 수면에 의한 영향 등과 같은 전파환경 변수들의 함수로 가정하고, 신경회로망 학습을 통하여 함수를 근사화한다. 전파환경 변수들이 신경회로망 입력으로 사용되기 위해서는 3차원 지형도와 벡터지도를 이용하여 전파의 반사, 회절, 산란 등의 물리적인 특성이 고려된 특징 추출을 통해 정량적인 수치들을 계산한다. 이와 같이 얻어진 훈련데이타를 이용한 신경회로망 학습을 통해 전파손실 모델을 완성한다. 이 모델을 이용하여 서울 도심 지역의 실제 서비스 환경에 대한 타 모델과의 비교실험결과를 통해 제안하는 모델의 우수성을 보인다.Abstract In cellular mobile communication systems, wave propagation models are used in most cases to predict cell coverage. The amount of propagation loss induced by the obstacles in the propagation path, however, is a highly non-linear function, which cannot be easily represented mathematically. In this paper, we introduce the method of producing propagation loss prediction models by function approximation using neural networks. In this method, we assume the propagation loss is a function of the relevant parameters such as the distance from the base station antenna, the specification of the transmitter antenna, obstacle profile, diffraction effect, road, and water effect. The values of these parameters are produced from the field measurement data, 3D digital terrain maps, and vector maps as its inputs by a feature extraction process, which takes into account the physical characteristics of electromagnetic waves such as reflection, diffraction and scattering. The values produced are used as the input to the neural network, which are then trained to become the propagation loss prediction model. In the experimental study, we obtain a considerable amount of improvement over COST-231 model in the prediction accuracy using this model.

• The Journal of the Korea Contents Association
• /
• v.6 no.1
• /
• pp.23-30
• /
• 2006

A clustering method for temporal data takes a model-based approach. This uses automata based model for each cluster. It is necessary to construct global models for a set of data in order to elicit individual models for the cluster. The preparation for building individual models is completed by determining the number of clusters inherent in the data set. In this paper, BIC(Bayesian Information Criterion) approximation is used to determine the number clusters and confirmed its applicability. A search technique to improve efficiency is also suggested by analyzing the relationship between data size and BIC values. A number of experiments have been performed to check its validity using artificially generated data sets. BIC approximation measure has been confirmed that it suggests best number of clusters through experiments provided that the number of data is relatively large.