• Journal of Multimedia Information System
• /
• v.8 no.3
• /
• pp.159-166
• /
• 2021

COVID-19 has dramatically changed people's daily life. Wearing masks is considered as a simple but effective way to defend the spread of the epidemic. Hence, a real-time and accurate mask wearing detection system is important. In this paper, a deep learning-based mask wearing detection system is developed to help people defend against the terrible epidemic. The system consists of three important functions, which are image detection, video detection and real-time detection. To keep a high detection rate, a deep learning-based method is adopted to detect masks. Unfortunately, according to the suddenness of the epidemic, the mask wearing dataset is scarce, so a mask wearing dataset is collected in this paper. Besides, to reduce the computational cost and runtime, a simple online and real-time tracking method is adopted to achieve video detection and monitoring. Furthermore, a function is implemented to call the camera to real-time achieve mask wearing detection. The sufficient results have shown that the developed system can perform well in the mask wearing detection task. The precision, recall, mAP and F1 can achieve 86.6%, 96.7%, 96.2% and 91.4%, respectively.

• Journal of Microbiology and Biotechnology
• /
• v.9 no.2
• /
• pp.184-190
• /
• 1999

The use of the polymerase chain reaction (PCR) method was described using two sets of primers based on the ceuN gene (JEJ 1 and JEJ 2) which encodes a protein involved in siderophore transport and 16S rRNA gene (pA and pB) for the sensitive and specific detection of enteropathogen Campylobacter jejuni. Six oligonucleotides were utilized in an amplification experiment and PCR products of predicted sizes were generated from whole cells and boiled cell lysates at the same intensity. Two sets of the primer pairs, JEJ and pAB, were specific enough for all C. jejuni strains tested for the direct use of whole cells without DNA extraction or lysis steps. In the PCR using the pAB primer pair, the detection limit, as determined by the ethidium bromide staining of the amplification products on agarose gels, was at the level of $10^1$ bacteria cells or less in both the pure culture and artificially inoculated milk and chicken enrichment samples, whereas the detection limit with the JEJ primer pair was relatively low, i.e. $10^3$ cells or more in the same PCR samples. The PCR method using either a primer JEJ or pAB was both repeatable and specific for the detection of C. jejuni in food. This method is simply completed within 4 h.

• Korean Journal of Veterinary Research
• /
• v.13 no.1
• /
• pp.39-46
• /
• 1973

Cultural method, dark field microscopy & fluorescent antibody technique were compared for their sensitivity of the detection of leptospires from experimentally infected mice. Two groups of mice were infected with L. icterohemorrhagiae (M20) and L. australis (Ballico), and the infected blood, urine and a number of organs were subjected to the bacterial isolation. The results obtained were summarized as follows: 1. L. icterohemorrhagiae (M20) and L. australis (Ballico) in blood, urine and various tissues of experimentally infected mice were detected with a negrigible non specificity, by the fluorescent antibody technique. 2. The fluorescent antibody technique, as applied to detection of leptospires in blood, urine and various infected tissue, proved to be better than cultural method and dark-field microscopy. 3. Early detection of leptospires by fluorescent antibody technique were possible in blood at 2 days after inoculation, whereas detection of organisms in liver, spleen, lung and kidney were observed later. By means of fluorescent antibody technique, the detection of leptospires in kidney and urine was possible up to 34 days postinoculation, whereas those in other parts were impossible. 4. Fluorescent antibody reaction of leptospires were highly specific to homologous antigen rather than to heterologous one. 5. Fluorescent antibody technique may be of value in the application for the demonstration of leptospira from clinical specimens.

• Journal of Convergence Society for SMB
• /
• v.3 no.1
• /
• pp.31-41
• /
• 2013

There have been various recent methods proposed in detecting driver drowsiness (DD) to avert fatal accidents. This work proposes a hardware/software (HW/SW) co-design approach in implementation of a DD detection system adapted from an AdaBoost-based object detection algorithm with Haar-like features [1] to monitor driver's eye closure rate. In this work, critical functions of the DD detection algorithm is accelerated through custom hardware components in order to speed up processing, while the software component implements the overall control and logical operations to achieve the complete functionality required of the DD detection algorithm. The HW/SW architecture was implemented on an Altera DE2 board with a video daughter board. Performance of the proposed implementation was evaluated and benchmarked against some recent works.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.28 no.1
• /
• pp.15-22
• /
• 1995

To study ecological properties of Vibrio cholerae non-Ol and Vibrio mimicus which have been described as new food poisoning bacteria recently, the influence of factors such as temperature, salinity, pH and chemical oxygen demand (COD) on detection rate and density of these bacteria were evaluated. Fifty four seawater samples and 49 bottom deposit samples from estuary of Kum river from March 26th, 1993 to February 22nd, 1994 were used for this study. The detection rate of V cholerae non-O1 were $16.7\%$ for seawater and $10.2\%$ for bottom deposit, respertively. The total detertion rate of V. cholerae non-O1 $(11.7\%)$ was a little higher than V mimicus $(10.7\%)$. Both V choierae non-O1 and V. mimicus were mainly detected in estuary water of which showed temperature $24^{\circ}C$ above and salinity $10\%o$ below. These bacteria were also detected in bottom deposit on January when the water temperature was $3.5^{\circ}C$. From these results, we supposed that temperature, salinity and organic material were important factors to growth of V. cholerae non-O1 and V. mimicus. V cholerae non-O1 might be grown better than V. mimicus under the fluctuating aquatic environmental condition such as salinity.

• Korean Journal of Veterinary Research
• /
• v.43 no.2
• /
• pp.239-246
• /
• 2003

Polymerase chain reaction (PCR) was evaluated for the early detection of Aujeszky's disease virus (ADV) DNA from virus-infected cell cultures. For the purposes, the Korean ADV NYJ1-87 was propagated in swine kidney (SK) cells and subjected to the amplification of DNA (217 bp) by PCR using sense and antisense primers specific to gp50 gene of the ADV. In detection of cell-associated viral DNA, reliable PCR conditions were determined as 30 cycles of reaction consisting 1 minute each of denaturation at $94^{\circ}C$, annealing at $55^{\circ}C$ and polymerization at $72^{\circ}C$. The PCR encountered best results with reagent mixtures of $50{\mu}l$ containing $200{\mu}M$ dNTPs, $0.2{\mu}M$ each sense and antisense primers, 1 mM $MgCl_2$ and 10% (v/v) template DNA in the final concentrations. ADV-specific DNAs were detected as early as 6, 6, and 9 hours post-infection, respectively, from lysates of the SK cells infected with ADV of $10^3$, $10^2$ and $10^1\;TCID_{50}/ml$ by this condition. In culture supernatant, the DNAs were detected from ADV of as low infectivity as $10^ {-3}\;TCID_{50}/ml$ by the reduced reagent concentrations and 30 cycles of 1 minute each of denaturation at $94^{\circ}C$ and annealing at $55^{\circ}C$, and 2 minutes of polymerization at $72^{\circ}C$. The lowest amount of detectable ADV DNA was 1 fg. In conclusion, the PCR condition established in the present study was recognized as a feasible alternative to time-consuming procedures in isolation and characterization of the virus.

• Bulletin of the Korean Chemical Society
• /
• v.27 no.2
• /
• pp.231-236
• /
• 2006

A novel sample pretreatment technique, headspace hanging drop liquid phase microextraction (HS-LPME) was studied and applied to the determination of flavors from solid clove buds by gas chromatography-mass spectrometry (GC-MS). Several parameters affecting on HS-LPME such as organic solvent drop volume, extraction time, extraction temperature and phase ratio were investigated. 1-Octanol was selected as the extracting solvent, drop size was fixed to 0.6 $\mu$L. 60 min extraction time at 25 ${^{\circ}C}$ was chosen. HS-LPME has the good efficiency demonstrated by the higher partition equilibrium constant ($K_{lh}$) values and concentration factor (CF) values. The limits of detection (LOD) were 1.5-3.2 ng. The amounts of eugenol, $\beta$-caryophyllene and eugenol acetate from the clove bud sample were 1.90 mg/g, 1.47 mg/g and 7.0 mg/g, respectively. This hanging drop based method is a simple, fast and easy sample enrichment technique using minimal solvent. HSLPME is an alternative sample preparation method for the analysis of volatile aroma compounds by GC-MS.

• Journal of Korean Society of Disaster and Security
• /
• v.10 no.2
• /
• pp.1-6
• /
• 2017

In the previous research, the research on the development of subway disaster detection system that discovers the disaster early warning of the subway station disaster and the evacuation to the passengers based on the Internet of things. This paper as a follow-up study analyzes the sensor data installed in the station in real time to quickly detect the disaster. In particular, we developed a statistical methodology based on the Mahalanobis distance in consideration of the environment that varies depending on the installation location of the sensor during initial system construction.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.5B
• /
• pp.489-499
• /
• 2003

This paper is a study about Honey-pot Model using CPN(Colored Petri Nets) that is a method of intrusion detection. Suggested Honey-pot model consists of two parts : \circled1 security kernel module for active induction of hacker's intrusion, intrusion detection and behavior pattern analysis. \circled2 virtual module for activity of induced hackers. However, suggested model was compared and analysed with conventional Denning model and Shieh nodel. The Honey-pot model using CPN can classify the characteristic of intrusion pattern, modeling intrusion pattern and pattern matching procedure, detect DDoS attack through multi hosts, and provide basis of study model for analysing intrusion pattern, finally.

• Bulletin of the Korean Chemical Society
• /
• v.31 no.11
• /
• pp.3077-3083
• /
• 2010

The electrochemical detection of As(III) was investigated on a platinum-iron(III) nanoparticles modified multiwalled carbon nanotube on glassy carbon electrode(nanoPt-Fe(III)/MWCNT/GCE) in 0.1 M $H_2SO_4$. The nanoPt-Fe(III)/MWCNT/GCE was prepared via continuous potential cycling in the range from -0.8 to 0.7 V (vs. Ag/AgCl), in 0.1 M KCl solution containing 0.9 mM $K_2PtCl_6$ and 0.6 mM $FeCl_3$. The Pt nanoparticles and iron oxide were co-electrodeposited into the MWCNT-Nafion composite film on GCE. The resulting electrode was examined by cyclic voltammetry (CV), scanning electron microscopy (SEM), and anodic stripping voltammetry (ASV). For the detection of As(III), the nanoPt-Fe(III)/MWCNT/GCE showed low detection limit of 10 nM (0.75 ppb) and high sensitivity of $4.76\;{\mu}A{\mu}M^{-1}$, while the World Health Organization's guideline value of arsenic for drinking water is 10 ppb. It is worth to note that the electrode presents no interference from copper ion, which is the most serious interfering species in arsenic detection.