• Korean Journal of Optics and Photonics
• /
• v.14 no.4
• /
• pp.383-391
• /
• 2003

A new optical security system for the protection of personal identification information is proposed. Personal identification information consisting of a pure face image and an identification number is used for verification and authentication. Image encryption is performed by a fully phase image encryption technique with two random phase masks located in the input and the Fourier plane of 4-f correlator. The personal information, however, can be leaked out in the decryption process. To cope with this possibility, the encrypted image itself is used in the identification process. An encrypted personal identification number is discriminated and recognized by using the proposed MMACE_p (multiplexed MACE_p) filter, and then authenticity of the personal information is verified by correlation of the face image using the optical wavelet matched filter (OWMF). MMACE_p filter is a synthetic filter with four MACE_p (minimum average correlation energy_phase encrypted) filters multiplexed in one filter plane to recognize 10 different encrypted-numbers at a time. OWMF can improve discrimination capability and SNR (signal to noise ratio). Computer simulations confirmed that the proposed security technique can be applied to the protection of personal identification information.

• Journal of Biosystems Engineering
• /
• v.41 no.1
• /
• pp.51-59
• /
• 2016

Purpose: This study examined the performance of two spectroscopy methods and multivariate classification methods to discriminate viable pepper seeds from their non-viable counterparts. Methods: A classification model for viable seeds was developed using partial least square discrimination analysis (PLS-DA) with Fourier transform near-infrared (FT-NIR) and Raman spectroscopic data in the range of $9080-4150cm^{-1}$ (1400-2400 nm) and $1800-970cm^{-1}$, respectively. The datasets were divided into 70% to calibration and 30% to validation. To reduce noise from the spectra and compare the classification results, preprocessing methods, such as mean, maximum, and range normalization, multivariate scattering correction, standard normal variate, and $1^{st}$ and $2^{nd}$ derivatives with the Savitzky-Golay algorithm were used. Results: The classification accuracies for calibration using FT-NIR and Raman spectroscopy were both 99% with first derivative, whereas the validation accuracies were 90.5% with both multivariate scattering correction and standard normal variate, and 96.4% with the raw data (non-preprocessed data). Conclusions: These results indicate that FT-NIR and Raman spectroscopy are valuable tools for a feasible classification and evaluation of viable pepper seeds by providing useful information based on PLS-DA and the threshold value.

• 한국지구물리탐사학회:학술대회논문집
• /
• 2006.06a
• /
• pp.9-14
• /
• 2006

Estimations of depth, magnetic orientation, and strength of dipole moments aid discrimination between unexploded ordnance (UXO) and non-UXO using magnetic surveys. Such estimations may be hindered by geologic noise, magnetic clutter, and overlapping tails of nearby dipole fields. An improved method of inversion for anomalies of single or multiple dipoles with arbitrary polarization was developed to include intra-inversion filtering and estimation of background field gradients. Data interpolated to grids are flagged so that only nodes nearest to measurement stations are used. To apply intra-inversion filtering to such data requires a gapped filter. Moreover, for data with significant gaps in coverage, or along the edges or corners of survey areas, intra-inversion filters must be appropriately modified. To that end, edge-adaptive and gapped gradient-nulling filters have been designed and tested. Applications are shown for magnetic field data from Chongcho Lake, Sokcho, Korea and the U. S. Army's Aberdeen Proving Ground in Maryland.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.16 no.4
• /
• pp.80-91
• /
• 2002

An on-line ultrasonic detector was developed to continuously monitor the ultrasonic signal due to partial discharge in transformer in service. The on-line ultrasonic detector has a band-pass filter designed to measure only the frequencies from 50 to 300[㎑] of ultrasonic signal, to remove electrical and mechanical noises from outside of the transformer, and tlle ultrasonic sensor contains a pre-amplifier with 60[dB] gain. The ultrasonic signal discrimination algorithm which discriminates the ultrasonic signal duration was developed to remove the ultrasonic signal due to OLTC operation having similar characteristics to those due to partial discharge. The reliability of the on-line ultrasonic detector developed in this study was convinced of measurement the ultrasonic signals from the model. transformer in laboratory and transformer in service.

• Korean Journal of Occupational Health Nursing
• /
• v.12 no.2
• /
• pp.146-155
• /
• 2003

This study was conducted for investigating the status of management of preemployment health examination and to have an effect on the worker's employment. Health managers of 103 companies in Incheon metropolitan city and Gyeonggi were interviewed by telephone. Of 103 companies, 67(65.1%) said they don't hire the applicants who have an active pulmonary tuberculosis, 80(77.7%) companies said they health HBV carrier is acceptable but active HBV carrier is not 29(28.2%) companies said they don't hire the applicants who have a hypertension or diabetes mellitus, 42(40.8%) companies said they don't hire the applicants who have a hearing disturbance. If HIVD is suspicious in X-ray lumbar-sacral region, 37(78.7% of 47 companies) said they do not hire the applicants. 29(35% of 83 companies) said they cancel the employment of the applicants who are suspicious of noise induced hearing loss on preplacement health examination. From our survey, preemployment health examination was utilizing mainly as a tool for the selection of health employees who don't have a disease. Furthermore, in many companies, additional test items are being included and getting more strict the selection criteria for preemployment health examination. For the right use of preemployment health examination, author suggested that further studies were needed to select the adequate test items and establish the reasonable criteria for preemployment health examination.

• Journal of the Korean Geophysical Society
• /
• v.2 no.2
• /
• pp.143-152
• /
• 1999

We have studied a denoising technique involving wavelet transform for improving the quality of geophysical data during the preprocessing stage. To assess the effectiveness of this technique, we have made synthetic data contaminated by random noises and compared the results of denoising with those obtained by conventional low-pass filtering. The low-pass filtering of the sinusoidal signal having a sharp discontinuity between the first and last sample values shows apparent errors related to Gibbs' phenomena. For the case of bump signal, the low-pass filtering induces maximum errors on peak values by removing some high-frequency components of signal itself. The wavelet transform technique, however, denoises these signals with much less adverse effects owing to its pertinent properties on locality of wavelet and easy discrimination of noise and signal in the wavelet domain. The field data of gravity tide are denoised by using soft threshold, which shrinked all the wavelet coefficients toward the origin, and the G-factor is determined by comparing the denoised data and theoretical data.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.24 no.2
• /
• pp.311-318
• /
• 2014

As modern computer graphics technology has been developed, it is hard to discriminate computer graphics from photographic images with the naked eye. Advances in graphics technology has brought a lot of convenience to human, it has side effects such as image forgery, malicious edit and fraudulent means. In order to cope with such problems, studies of various algorithms using a feature that represents a characteristic of an image has been processed. In this paper, we verify directly the existing algorithm, and provide new features based a noise that represents the characteristics of the computer graphics well. And this paper introduces the method of using SVM(Support Vector Machine) with features proposed in previous research to improve the discrimination accuracy.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.4
• /
• pp.159-164
• /
• 2023

Solar energy harvesting IoT devices prioritize maximizing the utilization of collected energy due to the periodic recharging nature of solar energy, rather than minimizing energy consumption. Meanwhile, research on edge AI, which performs machine learning near the data source instead of the cloud, is actively conducted for reasons such as data confidentiality and privacy, response time, and cost. One such research area involves performing various audio AI applications using audio data collected from multiple IoT devices in an IoT edge computing environment. However, in most studies, IoT devices only perform sensing data transmission to the edge server, and all processes, including data preprocessing, are performed on the edge server. In this case, it not only leads to overload issues on the edge server but also causes network congestion by transmitting unnecessary data for learning. On the other way, if data preprocessing is delegated to each IoT device to address this issue, it leads to another problem of increased blackout time due to energy shortages in the devices. In this paper, we aim to alleviate the problem of increased blackout time in devices while mitigating issues in server-centric edge AI environments by determining where the data preprocessed based on the energy state of each IoT device. In the proposed method, IoT devices only perform the preprocessing process, which includes sound discrimination and noise removal, and transmit to the server if there is more energy available than the energy threshold required for the basic operation of the device.

• Journal of Korea Society of Industrial Information Systems
• /
• v.29 no.2
• /
• pp.15-25
• /
• 2024

For effective analysis of animal ecosystems, technology that can automatically identify the current status of animal habitats is crucial. Specifically, animal sound classification, which identifies species based on their sounds, is gaining great attention where video-based discrimination is impractical. Traditional studies have relied on a single deep learning model to classify animal sounds. However, sounds collected in outdoor settings often include substantial background noise, complicating the task for a single model. In addition, data imbalance among species may lead to biased model training. To address these challenges, in this paper, we propose an animal sound classification scheme that combines predictions from multiple models using Focal Loss, which adjusts penalties based on class data volume. Experiments on public datasets have demonstrated that our scheme can improve recall by up to 22.6% compared to an average of single models.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.49 no.2
• /
• pp.94-101
• /
• 2012

This paper proposes an improved image steganalysis scheme to raise the detection rate of stego images out of cover images. To improve the detection rate of stego image in the steganalysis, tiny variation caused by data hiding should be amplified. For this, we extract feature vectors of cover image and stego image by two steps. First, we separate image into upper 4 bit subimage and lower 4 bit subimage. As a result, stego noise is expanded more than two times. We decompose separated subimages into twelve subbands by applying 3-level Haar wavelet transform and calculate co-occurrence probabilities of two different subbands in the same scale. Since co-occurrence probability of the two wavelet subbands is affected by data hiding, it can be used as a feature to differentiate cover images and stego images. The extracted feature vectors are used as the input to the multilayer perceptron(MLP) classifier to distinguish between cover and stego images. We test the performance of the proposed scheme over various embedding rates by the LSB, S-tool, COX's SS, and F5 embedding method. The proposed scheme outperforms the previous schemes in detection rate to existence of hidden message as well as exactness of discrimination.