• Journal of the Korea Society of Computer and Information
• /
• v.27 no.9
• /
• pp.21-32
• /
• 2022

Image labeling must be preceded in order to perform object detection, and this task is considered a significant burden in building a deep learning model. Tens of thousands of images need to be trained for building a deep learning model, and human labelers have many limitations in labeling these images manually. In order to overcome these difficulties, this study proposes a method to perform object detection without significant performance degradation, even though labeling some images rather than the entire image. Specifically, in this study, low-resolution oriental painting images are converted into high-quality images using a super-resolution algorithm, and the effect of SSIM and PSNR derived in this process on the mAP of object detection is analyzed. We expect that the results of this study can contribute significantly to constructing deep learning models such as image classification, object detection, and image segmentation that require efficient image labeling.

• Bulletin of the Korean Chemical Society
• /
• v.27 no.6
• /
• pp.835-840
• /
• 2006

Dimethyl 1-(4-nitrobenzoyl)-8-oxo-2,8-dihydro-1H-pyrazolo[5,1-a]isoindole-2,3-dicarboxylate has been used as an ionophore and o-nitrophenyloctyl ether as a plasticizer in order to develop a poly(vinyl chloride)-based membrane electrode for calcium ion detection. The sensors exhibit significantly enhanced response towards calcium(II) ions over the concentration range $8.0{\times}10^{-7}\;1.0{\times}10^{-1}$ M at pH 3.0-11 with a lower detection limit of $5.0 {\times}10^{-7}$ M. The sensors display Nernstian slope of 29.5 ${\pm}$ 0.5 mV per decade for Ca(II) ions. Effects of plasticizers, lipophilic salts and various foreign common ions are tested. It has a fast response time within 10 s over the entire concentration range and can be used for at least 2 months without any divergence in potentials. The proposed electrode revealed good selectivity and response for $Ca^{2+}$ over a wide variety of other metal ions. The selectivity of the sensor is comparable with those reported for other such electrodes. The proposed sensor was successfully applied as an indicator electrode for the potentiometric titration of a Ca(II) solution, with EDTA.

• Journal of the Korean Chemical Society
• /
• v.51 no.2
• /
• pp.147-153
• /
• 2007

A simple, selective and highly sensitive flow injection catalytic method was presented for determination of ruthenium based on its catalytic effect on the oxidation of pyronin B by periodate in pH=1.0. The reaction rate is controlled specrophotometricaly by monitoring the dye absorbance at 555 nm. The optimized conditions make it possible to determine ruthenium in the ranges of 0.1-10.0 ng/mL (r2=0.9982) and 10.0-50.0 ng/mL (r2=0.9934) with a detection limit of 0.04 ng/mL and a sample rate of 30±5 samples/h. Relative standard deviation for the results of five replicate measurements does not exceed 1.44%. The proposed method has been successfully applied for quantitation of ultra trace amounts of ruthenium in some environmental and biological samples.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.7
• /
• pp.3-9
• /
• 2013

This paper proposes a selective-weighted energy detection (S-WED) and a synchronization algorithm appropriate for it in IR-UWB system. Energy detectors that are practical in terms of implementation are employed widely for noncoherent reception in IR-UWB systems. However, they show low performance due to using the energy samples captured at symbol rate. For this reason, weighted energy detectors are developed to improve the performance of EDs. Hence, for WED, not only synchronization but also the weight coefficients are needed to be obtained prior to data detection. Meanwhile, the optimal weighting coefficients of WEDs are known to be energy values. Therefore, synchronization and the weighting coefficients can be obtained simultaneously. This paper proposes an S-WED and a simple synchronization algorithm for it in which sub-intervals having energies under a certain level are excluded in energy accumulation resulting in a simpler WED with a bit performance increase in low SNR region. The proposed algorithm is verified through simulations using the preamble symbol and channel models defined in the IEEE 802.15.4a.

• Journal of Sensor Science and Technology
• /
• v.23 no.3
• /
• pp.211-215
• /
• 2014

The Pt-, Ni- and Cr-decorated tubular $SnO_2$ nanofibers for gas sensors were prepared by the electrospinning of polyvinylpyrrolidone (PVP) nanofibers containing Pt, Ni, and Cr precursors, the sputtering of $SnO_2$ on the electrospun PVP nanofibers, and the removal of sacrificial PVP parts by heat treatment at $600^{\circ}C$ for 2 h. Pt-decorated tubular $SnO_2$ nanofibers showed high response ($R_a/R_g=210.5$, $R_g$: resistance in gas, $R_a$: resistance in air) to 5 ppm $C_2H_5OH$ at $350^{\circ}C$ with negligible cross-responses to other interference gases (5 ppm trimethylamine, $NH_3$, HCHO, p-xylene, toluene and benzene). Cr-decorated tubular $SnO_2$nanofibers showed the selective detection of p-xylene at $400^{\circ}C$. In contrast, no significant selectivity to a specific gas was found in Ni-decorated tubular $SnO_2$ nanofibers. The selective and sensitive detection of gases using Pt-decorated and Cr-decorated tubular $SnO_2$ nanofibers were discussed in relation to the catalytic promotion of gas sensing reaction.

• Applied Chemistry for Engineering
• /
• v.20 no.6
• /
• pp.571-580
• /
• 2009

Biochip sensing technologies offering in-situ, fast and real-time measurements in addition to portability can be powerfully utilized in a wide spectrum of research areas including environmental science, food science, medical diagnostics and drug development. In this article, we introduce current research trends and economic aspects of the development of various optical biochip technologies for the analysis of organophosphorus/carbamate pesticides in environmental samples, which is of global importance with serious consequences for both current and future generations. In particular, we will highlight recent efforts made in the creation of highly sensitive and selective optical biochip sensors in conjunction with nanobiotechnologies and microfabrication for the rapid detection of organophosphorus/carbamate pesticides.

• Journal of Adhesion and Interface
• /
• v.14 no.1
• /
• pp.13-20
• /
• 2013

Palladium plays a pivotal role in the production of dental and medicinal devices, medicinal substances, jewellery, automobile and high-performance adhesives. Despite the frequent and fruitful use of such reactions, one major setback is the high level of palladium in the resultant compounds which can harm the human body. Among the palladium species, $PdCl_2$ is the most toxic. As a consequence it is desirable to detect the $Pd^{2+}$ cations by fluorescence spectra because it can provide an operationally simple and cost-effective detection method together with high sensitivity and selectivity. Herein, an ${\alpha}$-carbonyl substituted pyrene derivative, ${\gamma}$-oxo-1-pyrenebutyric acid (OPBA), was demonstrated to be a highly sensitive and selective fluorescent probe for $Pd^{2+}$ among the metal cations examined in aqueous solutions.

• Bulletin of the Korean Chemical Society
• /
• v.30 no.7
• /
• pp.1497-1504
• /
• 2009

GC/MS analysis of catecholamines (CAs) in biological sample may produce poor reproducible quantitaion when chemical derivatization is used as the technique to form a volatile derivative. Significant quantities of the side products can be formed from CAs with primary amine during the derivatization reaction under un-optimized conditions. We have tested various chemical derivatization techniques in an attempt to find an optimum derivatization method that will reduce side product formation, enable to separate several catecholamine derivatives in GC chromatogram, and obtain significant improvement of detection sensitivity in GC/MS analysis. Whereas several derivatization techniques such as trimethylsilylation (TMS), trifluoroacylation (TFA), and two step derivatization methods were active, selective derivatization to form O-TMS, N-heptafluorobutylacyl (HFBA) derivative using N-methyl-N-(trimethylsilyl)-trifluoroacetamide (MSTFA) and N-methyl-bis(heptafluorobutyramide) (MBHFBA) reagents was found to be the most effective method. Moreover, this derivative formed by selective derivatization could provide sufficient sensitivity and peak separation as well as produce higher mass ion as base peak to use selected ion in SIM mode. Calibration curves based on the use of an isotopically labeled internal standard show good linearity over the range assayed, 1 ~ 5000 ng/mL, with correlation coefficients of > 0.996. The detection limits of the method ranged from 0.2 to 5.0 ppb for the different CAs studied. The developed method will be applied to the analysis of various CAs in biological sample, combined with appropriate sample pretreatment.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.1
• /
• pp.161-174
• /
• 1996

Addressed in this paper is performance analysis of the maximum-likelihood code acquisition technique for slotted-mode preamble search in the CDMA reverse link. The probabilities of detection, miss, and false alarm are derived analytically for a multiple $H_{1}$ cell case in a frequency-selective Rayleigh fading channel, based on the statics of the CDMA noncoherent demodulator output. the probability density function of the decision variable consisting of successive demodulator outputs is also derived by considering the fading characteristics of the received signal for both single and dual antenna cases. The performance of the code acquisition technique is evaluated numerically with an emphasis on investigating the effects of post-detection integration, fading rate, and antenna diversity on the detection performance.

• Bulletin of the Korean Chemical Society
• /
• v.33 no.5
• /
• pp.1608-1612
• /
• 2012

Singlet oxygen ($^1O_2$) is an important species for oxidation in biological processes. $^1O_2$ is implicated in the genotoxic effect, and plays an important role in the cell-signaling cascade and in the induction of gene expression. However, the rapid detection of $^1O_2$ in biological environments with sufficient specificity and sensitivity is hampered by its extremely low emission probability. Here, a layer-by-layer (LbL) film of CdTe quantum dots (QDs), polymers, and ascorbate have been designed as a rapid, highly selective, and sensitive fluorescence probe for $^1O_2$ detection. Upon reaction with $^1O_2$, the probe exhibits a strong photoluminescence (PL) response even at trace levels. This remarkable PL change should enable the probe to be used for $^1O_2$ detection in many chemical and biological systems and as an environmental sensor.