• Title/Summary/Keyword: recognition-rate

Search Result 2,809, Processing Time 0.035 seconds

A Comparative Study of Alzheimer's Disease Classification using Multiple Transfer Learning Models

  • Prakash, Deekshitha;Madusanka, Nuwan;Bhattacharjee, Subrata;Park, Hyeon-Gyun;Kim, Cho-Hee;Choi, Heung-Kook
    • Journal of Multimedia Information System
    • /
    • v.6 no.4
    • /
    • pp.209-216
    • /
    • 2019
  • Over the past decade, researchers were able to solve complex medical problems as well as acquire deeper understanding of entire issue due to the availability of machine learning techniques, particularly predictive algorithms and automatic recognition of patterns in medical imaging. In this study, a technique called transfer learning has been utilized to classify Magnetic Resonance (MR) images by a pre-trained Convolutional Neural Network (CNN). Rather than training an entire model from scratch, transfer learning approach uses the CNN model by fine-tuning them, to classify MR images into Alzheimer's disease (AD), mild cognitive impairment (MCI) and normal control (NC). The performance of this method has been evaluated over Alzheimer's Disease Neuroimaging (ADNI) dataset by changing the learning rate of the model. Moreover, in this study, in order to demonstrate the transfer learning approach we utilize different pre-trained deep learning models such as GoogLeNet, VGG-16, AlexNet and ResNet-18, and compare their efficiency to classify AD. The overall classification accuracy resulted by GoogLeNet for training and testing was 99.84% and 98.25% respectively, which was exceptionally more than other models training and testing accuracies.

Design and Implementation of a RFID Sensor Network Design System (RFID 센서 네트워크 설계 시스템의 설계 및 구현)

  • Soon, Nam-Soon;Yeo, Myung-Ho;Yoo, Jae-Soo
    • Proceedings of the Korea Contents Association Conference
    • /
    • 2009.05a
    • /
    • pp.505-509
    • /
    • 2009
  • Recently, many researchers have studied on RFID with the rapid development of RFID technologies. RFID adapts to many applications. A lot of readers may be deployed in some applications. The incorrect deployment of RFID readers occurs reader-to-reader interferences in many applications using RFID technologies. It causes the significant degradation of the RFID recognition rate. Therefore, it is very important to appropriately deploy the RFID readers in terms of costs and system performance. In this paper, we propose a novel RFID sensor network design system with the minimal RFID readers to cover whole regions when an arbitrary blueprint is given.

  • PDF

The Effects of Female Leadership on Job Stress and Promotion of Female Workers (여자의 적은 여자인가?: 상사 성별이 여성 근로자의 노동시장 성과에 미치는 영향 분석)

  • Jung, Han Na
    • Journal of Labour Economics
    • /
    • v.40 no.4
    • /
    • pp.61-89
    • /
    • 2017
  • The purpose of this study is to analyze the effect of sex of supervisor on the stress and promotion of Korean women workers in the workplace using the Korean Women Manager Panel. The results demonstrate that when the direct boss of female workers was female, stress in the workplace was high and the promotion rate of staff and assistant managers was low. According to the theory of queen bee phenomenon and role congruity theory, a structure of glass ceiling and the male-dominated working culture does not require positive qualities of female leadership but requires a high recognition standard or competitiveness. It can be interpreted that the female worker with the female supervisor showed more stress in the workplace and the probability of female promotion in staff and assistant managers is low.

  • PDF

Generation and characterization of calmodulin-DHFR sandwich fusion protein

  • Han, Chang Hoon
    • Korean Journal of Veterinary Research
    • /
    • v.48 no.3
    • /
    • pp.243-250
    • /
    • 2008
  • A calmodulin-dihydrofolate reductase (DHFR) sandwich fusion protein was generated by insertion of calmodulin into the $\beta$-bulge region of DHFR to observe the effects of structurally constraining the calmodulin structure. The calcium binding properties of the sandwich protein were almost identical to calmodulin. Similar to calmodulin ($10.7 {\mu}M$), the sandwich protein bound four equivalents of calcium, with half saturation ($K_{0.5}$) observed at a [$Ca^{2+}$] of $8{\mu}M$. However, nicotinamide adenine dinucleotide (NAD) kinase activation property of the sandwich protein was lower than that of calmodulin. The sandwich protein activated NAD kinase, but to only half of the level obtained with calmodulin. The K 0.5 for both calmodulin and the sandwich protein were approximately the same (1-2 nM). Methylation analyses of the sandwich protein show that insertion of calmodulin into DHFR results in a large decrease in methylation. The $V_{max}$ observed with the sandwich protein (95 nmole/min/ml) was only 22% of the value observed with calmodulin (436 nmol/min/ml) in the presence of calcium. Addition of trimethoprim to the reaction significantly inhibited the observed methylation rate. Overall, the data suggest that the insertion of calmodulin into the DHFR structure has little effect on calcium binding by the individual lobes of calmodulin, but may constrain the lobes in a manner that results in altered interaction with the calmodulin-dependent proteins, and severely perturbed the methyltransferase recognition site.

Prevalence Rate and Factors Associated with Atopic Dermatitis among Korean Middle School Students

  • Oak, Ji Won;Lee, Hea Shoon
    • Journal of Korean Academy of Nursing
    • /
    • v.42 no.7
    • /
    • pp.992-1000
    • /
    • 2012
  • Purpose: The purpose of this study was to verify prevalence rates of atopic dermatitis (AD) in middle school students in Korea and to define AD and associated factors. Methods: This study was conducted using secondary analysis of raw data from the 'The 6th Korean National Adolescent Health Behavior Survey, 2010' conducted by the Korea Centers for Disease Control. The participants were 37,570 students. General characteristics and health behavior-related characteristics were analyzed using descriptive statistics, ${\chi}^2$-test and logistic regression analysis. Results: Results are as follows. 1. There were 8,494 middle school students (22.6%), who had been diagnosed with AD during the last 12 months. 2. Factors associated with AD in middle school students were gender (p<.001), grade level (p=.001), father's education (p<.001), mother's education (p=.032), regular dietary habits (breakfasts, p=.025; lunches; p=.003), stress (p<.001), recognition of subjective awareness of one's health conditions (p<.001), hand washing before meals at school (p=.013), and cessation education for both smoking (p=.002) and alcohol (p=.031). Conclusion: Results of the study indicate that in order to alleviate AD symptoms in these students, there is a need to develop nursing interventions and education programs taking into consideration these variables and verifying the effects of interventions and programs.

Development of Data Fusion Human Identification System Based on Finger-Vein Pattern-Matching Method and photoplethysmography Identification

  • Ko, Kuk Won;Lee, Jiyeon;Moon, Hongsuk;Lee, Sangjoon
    • International Journal of Internet, Broadcasting and Communication
    • /
    • v.7 no.2
    • /
    • pp.149-154
    • /
    • 2015
  • Biometric techniques for authentication using body parts such as a fingerprint, face, iris, voice, finger-vein and also photoplethysmography have become increasingly important in the personal security field, including door access control, finance security, electronic passport, and mobile device. Finger-vein images are now used to human identification, however, difficulties in recognizing finger-vein images are caused by capturing under various conditions, such as different temperatures and illumination, and noise in the acquisition camera. The human photoplethysmography is also important signal for human identification. In this paper To increase the recognition rate, we develop camera based identification method by combining finger vein image and photoplethysmography signal. We use a compact CMOS camera with a penetrating infrared LED light source to acquire images of finger vein and photoplethysmography signal. In addition, we suggest a simple pattern matching method to reduce the calculation time for embedded environments. The experimental results show that our simple system has good results in terms of speed and accuracy for personal identification compared to the result of only finger vein images.

Estimating Population Density of Leopard Cat (Prionailurus bengalensis) from Camera Traps in Maekdo Riparian Park, South Korea

  • Park, Heebok;Lim, Anya;Choi, Tae-Young;Lim, Sang-Jin;Park, Yung-Chul
    • Journal of Forest and Environmental Science
    • /
    • v.33 no.3
    • /
    • pp.239-242
    • /
    • 2017
  • Although camera traps have been widely used to understand the abundance of wildlife in recent decades, the effort has been restricted to small sub-set of wildlife which can mark-and-recapture. The Random Encounter Model shows an alternative approach to estimate the absolute abundance from camera trap detection rate for any animals without the need for individual recognition. Our study aims to examine the feasibility and validity of the Random Encounter Model for the density estimation of endangered leopard cats (Prionailurus bengalensis) in Maekdo riparian park, Busan, South Korea. According to the model, the estimated leopard cat density was $1.76km^{-2}$ (CI 95%, 0.74-3.49), which indicated 2.46 leopard cats in $1.4km^2$ of our study area. This estimate was not statistically different from the previous leopard cat population count ($2.33{\pm}0.58$) in the same area. As follows, our research demonstrated the application and usefulness of the Random Encounter Model in density estimation of unmarked wildlife which helps to manage and protect the target species with a better understanding of their status.

Fall detection of the elderly through floor vibrations (바닥 진동을 통한 노인 낙상 검출)

  • Kim, Dong-Wan;Ryu, Jong-Hyun;Beack, Seung-Hwa
    • Journal of IKEEE
    • /
    • v.18 no.1
    • /
    • pp.134-139
    • /
    • 2014
  • According to survey, more than 57.2% of the fall which is the most frequent safety accident of the elders takes place at home. This research aims to verify the fall by measuring and analyzing the floor vibration. And the vibration sensor module was designed with piezo film sensor and operation amplifier. The vibration signals are converted to digital signals through the data acquisition device and vibration sensor module. And then modified the signals into frequency domain to obtain characteristic vibration data. The characteristic signals are verified by K-Nearest Neighbor verification, and experimental results shows the recognition rate 93.6%. Also the fall detection sensor module is useful for extract the meaningful data for fall detection. 10 persons are participated for this experiment.

Development of a sdms (Self-diagnostic monitoring system) with prognostics for a reciprocating pump system

  • Kim, Wooshik;Lim, Chanwoo;Chai, Jangbom
    • Nuclear Engineering and Technology
    • /
    • v.52 no.6
    • /
    • pp.1188-1200
    • /
    • 2020
  • In this paper, we consider a SDMS (Self-Diagnostic Monitoring System) for a reciprocating pump for the purpose of not only diagnosis but also prognosis. We have replaced a multi class estimator that selects only the most probable one with a multi label estimator such that we are able to see the state of each of the components. We have introduced a measure called certainty so that we are able to represent the symptom and its state. We have built a flow loop for a reciprocating pump system and presented some results. With these changes, we are not only able to detect both the dominant symptom as well as others but also to monitor how the degree of severity of each component changes. About the dominant ones, we found that the overall recognition rate of our algorithm is about 99.7% which is slightly better than that of the former SDMS. Also, we are able to see the trend and to make a base to find prognostics to estimate the remaining useful life. With this we hope that we have gone one step closer to the final goal of prognosis of SDMS.

Detection of eye using optimal edge technique and intensity information (눈 영역에 적합한 에지 추출과 밝기값 정보를 이용한 눈 검출)

  • Mun, Won-Ho;Choi, Yeon-Seok;Kim, Cheol-Ki;Cha, Eui-Young
    • Proceedings of the Korean Institute of Information and Commucation Sciences Conference
    • /
    • 2010.10a
    • /
    • pp.196-199
    • /
    • 2010
  • The human eyes are important facial landmarks for image normalization due to their relatively constant interocular distance. This paper introduces a novel approach for the eye detection task using optimal segmentation method for eye representation. The method consists of three steps: (1)edge extraction method that can be used to accurately extract eye region from the gray-scale face image, (2)extraction of eye region using labeling method, (3)eye localization based on intensity information. Experimental results show that a correct eye detection rate of 98.9% can be achieved on 2408 FERET images with variations in lighting condition and facial expressions.

  • PDF