• ETRI Journal
• /
• v.46 no.3
• /
• pp.513-525
• /
• 2024

Deep neural networks trained on labeled medical data face major challenges owing to the economic costs of data acquisition through expensive medical imaging devices, expert labor for data annotation, and large datasets to achieve optimal model performance. The heterogeneity of diseases, such as Alzheimer's disease, further complicates deep learning because the test cases may substantially differ from the training data, possibly increasing the rate of false positives. We propose a reconstruction-based self-supervised anomaly detection model to overcome these challenges. It has a dual-subnetwork encoder that enhances feature encoding augmented by skip connections to the decoder for improving the gradient flow. The novel encoder captures local and global features to improve image reconstruction. In addition, we introduce an entropy-based image conversion method. Extensive evaluations show that the proposed model outperforms benchmark models in anomaly detection and classification using an encoder. The supervised and unsupervised models show improved performances when trained with data preprocessed using the proposed image conversion method.

• Communications for Statistical Applications and Methods
• /
• v.6 no.3
• /
• pp.909-918
• /
• 1999

Arizono and Ohta(1989) studied goodness of fit test of normality using the entropy estimator proposed by Vasicek (1976) Recently van Es(1992) and Correa(1995) proposed an estimator of entropy. In this paper we propose goodness of fit test statistics for normality based on Vasicek ven Es and Correa. And we compare the power of the proposed test statistics with Kolmogorov-Smirnov Kuiper Cramer von Mises Watson Anderson-Darling and Finkelstein and Schefer statistics.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.4
• /
• pp.991-998
• /
• 2010

In realization of wireless home network system using speech recognition in indoor voice recognition environment, background noise and reverberation are two main causes of digression in voice recognition system. In this study, the home network system resistant to reverberation and background noise using voice section detection method based on spectral entropy in indoor recognition environment is to be realized. Spectral subtraction can reduce the effect of reverberation and remove noise independent from voice signal by eliminating signal distorted by reverberation in spectrum. For effective spectral subtraction, the correct separation of voice section and silent section should be accompanied and for this, improvement of performance needs to be done, applying to voice section detection method based on entropy. In this study, experimental and indoor environment testing is carried out to figure out command recognition rate in indoor recognition environment. The test result shows that command recognition rate improved in static environment and reverberant room condition, using voice section detection method based on spectral entropy.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.10
• /
• pp.2211-2217
• /
• 2009

In this paper, we propose the speech enhancement algorithm using an one-microphone in a reverberant room environments. Spectral subtraction is the effective method which can reduce the reverberation element and the noise in a spectrum domain. Spectral subtraction needs correct separation of voice section and silent section therefore to improve the performance, voice activity detection(VAD) based on entropy has been applied to the proposed method. We test a performance of the proposed method by comparing with conventional method which used VAD based on energy detection. Reverberation reduction ratio with variable of SNR and a reverberation time is used as a test index. From the simulation result, proposed method shows performance better than conventional method.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.13 no.1
• /
• pp.115-125
• /
• 2021

Simulation-based hull form optimization is a typical HEB (high-dimensional, expensive computationally, black-box) problem. Conventional optimization algorithms easily fall into the "curse of dimensionality" when dealing with HEB problems. A recently proposed Cross-Entropy (CE) optimization algorithm is an advanced stochastic optimization algorithm based on a probability model, which has the potential to deal with high-dimensional optimization problems. Currently, the CE algorithm is still in the theoretical research stage and rarely applied to actual engineering optimization. One reason is that the Monte Carlo (MC) method is used to estimate the high-dimensional integrals in parameter update, leading to a large sample size. This paper proposes an improved CE algorithm based on quasi-Monte Carlo (QMC) estimation using high-dimensional truncated Sobol subsequence, referred to as the QMC-CE algorithm. The optimization performance of the proposed algorithm is better than that of the original CE algorithm. With a set of identical control parameters, the tests on six standard test functions and a hull form optimization problem show that the proposed algorithm not only has faster convergence but can also apply to complex simulation optimization problems.

• The Korean Journal of Applied Statistics
• /
• v.24 no.6
• /
• pp.1271-1284
• /
• 2011

The entropy-based test of fit for the inverse Gaussian distribution presented by Mudholkar and Tian(2002) can only be applied to the composite hypothesis that a sample is drawn from an inverse Gaussian distribution with both the location and scale parameters unknown. In application, however, a researcher may want a test of fit either for an inverse Gaussian distribution with one parameter known or for an inverse Gaussian distribution with both the two partameters known. In this paper, we introduce tests of fit for the inverse Gaussian distribution based on the Kullback-Leibler information as an extension of the entropy-based test. A window size should be chosen to implement the proposed tests. By means of Monte Carlo simulations, window sizes are determined for a wide range of sample sizes and the corresponding critical values of the test statistics are estimated. The results of power analysis for various alternatives report that the Kullback-Leibler information-based goodness-of-fit tests have good power.

• Journal of Information Processing Systems
• /
• v.11 no.2
• /
• pp.248-265
• /
• 2015

In this paper, we propose a maximum entropy-based model, which can mathematically explain the bio-molecular event extraction problem. The proposed model generates an event table, which can represent the relationship between an event trigger and its arguments. The complex sentences with distinctive event structures can be also represented by the event table. Previous approaches intuitively designed a pipeline system, which sequentially performs trigger detection and arguments recognition, and thus, did not clearly explain the relationship between identified triggers and arguments. On the other hand, the proposed model generates an event table that can represent triggers, their arguments, and their relationships. The desired events can be easily extracted from the event table. Experimental results show that the proposed model can cover 91.36% of events in the training dataset and that it can achieve a 50.44% recall in the test dataset by using the event table.

• Journal of the Korean Data and Information Science Society
• /
• v.26 no.6
• /
• pp.1573-1582
• /
• 2015

In lifetime data analysis, it is generally known that the lifetimes of test items may not be recorded exactly. There are also situations wherein the withdrawal of items prior to failure is prearranged in order to decrease the time or cost associated with experience. Moreover, it is generally known that more than one cause or risk factor may be present at the same time. Therefore, analysis of censored competing risks data are needed. In this article, we derive the Bayes estimators for the entropy function under the exponential distribution with an unknown scale parameter based on multiply Type II censored competing risks data. The Bayes estimators of entropy function for the exponential distribution with multiply Type II censored competing risks data under the squared error loss function (SELF), precautionary loss function (PLF) and DeGroot loss function (DLF) are provided. Lindley's approximate method is used to compute these estimators.We compare the proposed Bayes estimators in the sense of the mean squared error (MSE) for various multiply Type II censored competing risks data. Finally, a real data set has been analyzed for illustrative purposes.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.27 no.3
• /
• pp.449-457
• /
• 2017

In this paper, we analyze the security and performance of the Quantis Quantum random number generator in terms of cryptography through experiments. The Quantis' post-processing is designed to output full-entropy via bit-matrix-vector multiplication based on mathematical background, and we used the min-entropy estimating test of NIST SP 800-90B so as to verify whether the output is full-entropy. Quantis minimizes the effect on the random bit rate by using an optimization technique for bit-matrix-vector multiplication, and compared the performance to conditioning functions of NIST SP 800-90B by measuring the random bit rate. Also, we have distinguished what is in Quantis' post-processing to the standard model of NIST in USA and BSI in Germany, and in case of applying Quantis to cryptographic systems in accordance with the CMVP standard, it is recommended to use the output of Quantis as the seed of the approved DRBG.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.1
• /
• pp.183-189
• /
• 2013

This paper proposes a novel lossless image compression scheme composed of direction-adaptive prediction and context-based entropy coding. In the prediction stage, we analyze the directional property with respect to the current coding pixel and select an appropriate prediction pixel. In order to further reduce the prediction error, we propose a prediction error compensation technique based on the context model defined by the activities and directional properties of neighboring pixels. The proposed scheme applies a context-based Golomb-Rice coding as the entropy coding since the coding efficiency can be improved by using the conditional entropy from the viewpoint of the information theory. Experimental results indicate that the proposed lossless image compression scheme outperforms the low complexity and high efficient JPEG-LS in terms of the coding efficiency by 1.3% on average for various test images, specifically for the images with a remarkable direction the proposed scheme shows better results.