• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.49 no.2
• /
• pp.113-121
• /
• 2012

In this paper, we proposed fast and efficient algorithm for detecting near-duplication based on content based retrieval in large scale video database. For handling large amounts of video easily, we split the video into small segment using scene change detection. In case of video services and copyright related business models, it is need to technology that detect near-duplicates, that longer matched video than to search video containing short part or a frame of original. To detect near-duplicate video, we proposed motion distribution and frame descriptor in a video segment. The motion distribution descriptor is constructed by obtaining motion vector from macro blocks during the video decoding process. When matching between descriptors, we use the motion distribution descriptor as filtering to improving matching speed. However, motion distribution has low discriminability. To improve discrimination, we decide to identification using frame descriptor extracted from selected representative frames within a scene segmentation. The proposed algorithm shows high success rate and low false alarm rate. In addition, the matching speed of this descriptor is very fast, we confirm this algorithm can be useful to practical application.

• Korean Journal of Veterinary Service
• /
• v.37 no.4
• /
• pp.263-271
• /
• 2014

In this study, we developed a cost and time saving one-step multiplex RT-PCR for the simultaneous detection and differentiation of swine influenza viruses (SIV) and 2009 pandemic influenza H1N1 virus (pH1N1). The one-step multiplex RT-PCR using four sets of primer was confirmed to be capable of detection of all SIV subtypes and differential diagnosis of major SIV subtype H1, H3 and pH1N1 on individual or mixed viral culture samples. The sensitivity of the multiplex RT-PCR was determined to be at least $2^{-6}$ $HA/25{\mu}L$ of the presented SIVs, providing sufficient efficacy for a routine SIV monitoring in diagnostic laboratories. In addition, compared with the conventional RT-PCR methods that cannot avoid the carryover DNA contamination, the developed RT-PCR applied with the uracil DNA glycosylase (UNG) system was proven to prevent a false positive reaction by carryover contamination of the pre-amplified DNA. In conclusion, the one-step RT-PCR with UNG system could be applicable to detect and differentiate of SIV from the viral cultures without worry of carryover DNA contamination in clinical laboratories.

• Korean Journal of Food Science and Technology
• /
• v.44 no.6
• /
• pp.673-679
• /
• 2012

Ten commercially available red pepper powders were investigated using photostimulated-luminescence (PSL), thermoluminescence (TL) and electron spin resonance (ESR) analyses to confirm their irradiation status. The application of PSL, TL, and ESR analyses was also confirmed by in-house irradiation. In PSL-based screening, all samples gave negative photon counts (<700 PCs). The PSL calibration dose (1 kGy) showed a low sensitivity of 4 samples, while the others provided reliable screening results. TL glow curves demonstrated maximum peaks after $250^{\circ}C$ for the 6 samples; however 4 samples gave complex TL glow curves with maximum peaks in the range of $185-260^{\circ}C$ (radiation-specific), which could be the effect of an irradiated component in low concentration as the TL ratios of all samples were <0.1. Radiation-specific ESR features were absent in the all commercial samples. Variable irradiation detection properties were found; where the TL analysis showed the possible presence of an irradiated component in 4 samples requiring further monitoring and investigation.

• Journal of Breast Cancer
• /
• v.21 no.4
• /
• pp.453-462
• /
• 2018

Purpose: This study aimed to compare the diagnostic performance of contrast-enhanced digital mammography (CEDM) and contrast-enhanced magnetic resonance imaging (CEMRI) in preoperative evaluations, and to evaluate the effect of each modality on the surgical management of women with breast cancer. Methods: This single-center, prospective study was approved by the Institutional Review Board, and informed consent was obtained from all patients. From November 2016 to October 2017, 84 patients who were diagnosed with invasive carcinoma (69/84) and ductal carcinoma in situ (15/84), and underwent both CEDM and CEMRI, were enrolled. Imaging findings and surgical management were correlated with pathological results and compared. The diagnostic performance of both modalities in the detection of index and secondary cancers (multifocality and multicentricity), and occult cancer in the contralateral breast, was compared. The authors also evaluated whether CEDM or CEMRI resulted in changes in the surgical management of the affected breast due to imaging-detected findings. Results: Eighty-four women were included in the analysis. Compared with CEMRI, CEDM demonstrated a similar sensitivity (92.9% [78/84] vs. 95.2% [80/84]) in detecting index cancer (p=0.563). For the detection of secondary cancers in the ipsilateral breast and occult cancer in the contralateral breast, no significant differences were found between CEDM and CEMRI (p=0.999 and p=0.999, respectively). Regarding changes in surgical management, CEDM resulted in similar changes compared with CEMRI (30.9% [26/84] vs. 29.7% [25/84], p=0.610). Regarding changes in surgical management due to false-positive findings, no significant differences were found between CEDM and CEMRI (34.6% [9/26] vs. 44.0% [11/25], p=0.782). Conclusion: CEDM demonstrated a diagnostic performance comparable with CEMRI in depicting index cancers, secondary cancers, and occult cancer in the contralateral breast. CEDM demonstrated similar changes in surgical management compared with CEMRI.

• The Journal of the Korea Contents Association
• /
• v.21 no.1
• /
• pp.531-540
• /
• 2021

As cyber attacks and crimes increase exponentially and hacking attacks become more intelligent and advanced, hacking attack methods and routes are evolving unpredictably and in real time. In order to reinforce the enemy's responsiveness, this study aims to propose a method for developing an artificial intelligence-based security control platform by building a next-generation security system using artificial intelligence to respond by self-learning, monitoring abnormal signs and blocking attacks.The artificial intelligence-based security control platform should be developed as the basis for data collection, data analysis, next-generation security system operation, and security system management. Big data base and control system, data collection step through external threat information, data analysis step of pre-processing and formalizing the collected data to perform positive/false detection and abnormal behavior analysis through deep learning-based algorithm, and analyzed data Through the operation of a security system of prevention, control, response, analysis, and organic circulation structure, the next generation security system to increase the scope and speed of handling new threats and to reinforce the identification of normal and abnormal behaviors, and management of the security threat response system, Harmful IP management, detection policy management, security business legal system management. Through this, we are trying to find a way to comprehensively analyze vast amounts of data and to respond preemptively in a short time.

• The Korean Journal of Applied Statistics
• /
• v.36 no.2
• /
• pp.101-114
• /
• 2023

The data augmentation technique is effectively used to solve the problem of overfitting the model by allowing the training dataset to be viewed from various perspectives. In addition to image augmentation techniques such as rotation, cropping, horizontal flip, and vertical flip, occlusion-based data augmentation methods such as Cutmix and Cutout have been proposed. For models based on speech data, it is possible to use an occlusion-based data-based augmentation technique after converting a 1D speech signal into a 2D spectrogram. In particular, SpecAugment is an occlusion-based augmentation technique for speech spectrograms. In this study, we intend to compare and study data augmentation techniques that can be used in the problem of false-voice detection. Using data from the ASVspoof2017 and ASVspoof2019 competitions held to detect fake audio, a dataset applied with Cutout, Cutmix, and SpecAugment, an occlusion-based data augmentation method, was trained through an LCNN model. All three augmentation techniques, Cutout, Cutmix, and SpecAugment, generally improved the performance of the model. In ASVspoof2017, Cutmix, in ASVspoof2019 LA, Mixup, and in ASVspoof2019 PA, SpecAugment showed the best performance. In addition, increasing the number of masks for SpecAugment helps to improve performance. In conclusion, it is understood that the appropriate augmentation technique differs depending on the situation and data.

• Korean Journal of Poultry Science
• /
• v.44 no.2
• /
• pp.93-102
• /
• 2017

A duplex RT-PCR (dRT-PCR) assay was developed for the simultaneous detection and discrimination of non-virulent and virulent Newcastle disease virus (NDV) in a single PCR tube. Primers targeting the large polymerase protein (L) gene and the fusion protein (F) gene of NDV were designed to detect all NDVs (by common type PCR primers) and virulent NDVs (by pathotype PCR primers), respectively and evaluated experimentally with reference NDV strains and other poultry viral pathogens. PCR products of the expected size of 386 bp were amplified from all NDV samples whereas PCR products of the expected size of 229 bp were amplified from virulent NDV samples alone. Cross reaction was not observed with other avian viral pathogens. The detection limit of NDV by the dRT-PCR was estimated to be $10^3$ 50% egg infectious dose/0.1 mL. In the dRT-PCR using field isolates of NDV, the pathotype PCR primers detected specifically all of virulent field isolates of NDV from Malaysia, Pakistan and China whereas common type PCR primers detected 94.4% (51/54) of field isolates of NDV from China. Three Chinese NDV isolates with false negative result were non-virulent viruses. Our results indicate that the dRT-PCR might provide a rapid and simple tool for rapid simultaneous detection and discrimination of non-virulent and virulent NDVs. Therefore the developed dRT-PCR assay provides a powerful novel means for the rapid diagnosis of Newcastle disease.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_1
• /
• pp.1353-1369
• /
• 2023

In the event of a disaster accident at sea, the scale of damage will vary due to weather effects such as wind, currents, and tidal waves, and it is obligatory to minimize the scale of damage by establishing appropriate control plans through quick on-site identification. In particular, it is difficult to identify pollutants that exist in a thin film at sea surface due to their relatively low viscosity and surface tension among pollutants discharged into the sea. Therefore, this study aims to develop an algorithm to detect suspended pollutants on the sea surface in RGB images using imaging equipment that can be easily used in the field, and to evaluate the performance of the algorithm using input data obtained from actual waters. The developed algorithm uses image enhancement techniques to improve the contrast between the intensity values of pollutants and general sea surfaces, and through histogram analysis, the background threshold is found,suspended solids other than pollutants are removed, and finally pollutants are classified. In this study, a real sea test using substitute materials was performed to evaluate the performance of the developed algorithm, and most of the suspended marine pollutants were detected, but the false detection area occurred in places with strong waves. However, the detection results are about three times better than the detection method using a single threshold in the existing algorithm. Through the results of this R&D, it is expected to be useful for on-site control response activities by detecting suspended marine pollutants that were difficult to identify with the naked eye at existing sites.

• Journal of the Korean Society of Radiology
• /
• v.81 no.4
• /
• pp.886-898
• /
• 2020

Purpose The purpose of our study was to evaluate digital breast tomosynthesis as a breast cancer screening modality for women with gynecologic cancer. Materials and Methods This retrospective study included patients with underlying gynecologic malignancies who underwent screening digital breast tomosynthesis for breast cancer. The cancer detection rate, recall rate, sensitivity, specificity, and positive predictive value (PPV) were calculated. PPV1 was defined as the percentage of all positive screening exams that have a tissue diagnosis of cancer within a year. PPV2 was defined as the percentage of all diagnostic exams (and Breast Imaging Reporting and Data System category 4, 5 from screening setting) with a recommendation for tissue diagnosis that have cancer within a year. PPV3 was defined as the percentage of all known biopsies actually performed that resulted in a tissue diagnosis of cancer within the year. For each case of screen-detected cancer, we analyzed the age, type of underlying gynecologic malignancy, breast density, imaging features, final Breast Imaging Reporting and Data System assessment, histologic type, T and N stages, molecular subtype, and Ki-67 index. Results Among 508 patients, 7 with breast cancer were identified after a positive result. The cancer detection rate was 13.8 per 1000 screening exams, and the recall rate was 17.9%. The sensitivity was 100%, and the specificity was 83.2%. The false negative rate was 0 per 1000 exams. The PPV1, PPV2, and PPV3 were 7.7, 31.8, and 31.8, respectively. Conclusion Digital breast tomosynthesis may be a promising breast cancer screening modality for women with gynecologic cancer, based on the high cancer detection rate, high sensitivity, high PPV, and high detection rate of early-stage cancer observed in our study.

• Journal of the Korean earth science society
• /
• v.33 no.2
• /
• pp.139-147
• /
• 2012

Mongolia's solar-meteorological resources map has been developed using satellite data and reanalysis data. Solar radiation was calculated using solar radiation model, in which the input data were satellite data from SRTM, TERA, AQUA, AURA and MTSAT-1R satellites and the reanalysis data from NCEP/NCAR. The calculated results are validated by the DSWRF (Downward Short-Wave Radiation Flux) from NCEP/NCAR reanalysis. Mongolia is composed of mountainous region in the western area and desert or semi-arid region in middle and southern parts of the country. South-central area comprises inside the continent with a clear day and less rainfall, and irradiation is higher than other regions on the same latitude. The western mountain region is reached a lot of solar energy due to high elevation but the area is covered with snow (high albedo) throughout the year. The snow cover is a cause of false detection from the cloud detection algorithm of satellite data. Eventually clearness index and solar radiation are underestimated. And southern region has high total precipitable water and aerosol optical depth, but high solar radiation reaches the surface as it is located on the relatively lower latitude. When calculated solar radiation is validated by DSWRF from NCEP/NCAR reanalysis, monthly mean solar radiation is 547.59 MJ which is approximately 2.89 MJ higher than DSWRF. The correlation coefficient between calculation and reanalysis data is 0.99 and the RMSE (Root Mean Square Error) is 6.17 MJ. It turned out to be highest correlation (r=0.94) in October, and lowest correlation (r=0.62) in March considering the error of cloud detection with melting and yellow sand.