• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.1-25
• /
• 2020

In this paper, we suggest an application system architecture which provides accurate, fast and efficient automatic gasometer reading function. The system captures gasometer image using mobile device camera, transmits the image to a cloud server on top of private LTE network, and analyzes the image to extract character information of device ID and gas usage amount by selective optical character recognition based on deep learning technology. In general, there are many types of character in an image and optical character recognition technology extracts all character information in an image. But some applications need to ignore non-of-interest types of character and only have to focus on some specific types of characters. For an example of the application, automatic gasometer reading system only need to extract device ID and gas usage amount character information from gasometer images to send bill to users. Non-of-interest character strings, such as device type, manufacturer, manufacturing date, specification and etc., are not valuable information to the application. Thus, the application have to analyze point of interest region and specific types of characters to extract valuable information only. We adopted CNN (Convolutional Neural Network) based object detection and CRNN (Convolutional Recurrent Neural Network) technology for selective optical character recognition which only analyze point of interest region for selective character information extraction. We build up 3 neural networks for the application system. The first is a convolutional neural network which detects point of interest region of gas usage amount and device ID information character strings, the second is another convolutional neural network which transforms spatial information of point of interest region to spatial sequential feature vectors, and the third is bi-directional long short term memory network which converts spatial sequential information to character strings using time-series analysis mapping from feature vectors to character strings. In this research, point of interest character strings are device ID and gas usage amount. Device ID consists of 12 arabic character strings and gas usage amount consists of 4 ~ 5 arabic character strings. All system components are implemented in Amazon Web Service Cloud with Intel Zeon E5-2686 v4 CPU and NVidia TESLA V100 GPU. The system architecture adopts master-lave processing structure for efficient and fast parallel processing coping with about 700,000 requests per day. Mobile device captures gasometer image and transmits to master process in AWS cloud. Master process runs on Intel Zeon CPU and pushes reading request from mobile device to an input queue with FIFO (First In First Out) structure. Slave process consists of 3 types of deep neural networks which conduct character recognition process and runs on NVidia GPU module. Slave process is always polling the input queue to get recognition request. If there are some requests from master process in the input queue, slave process converts the image in the input queue to device ID character string, gas usage amount character string and position information of the strings, returns the information to output queue, and switch to idle mode to poll the input queue. Master process gets final information form the output queue and delivers the information to the mobile device. We used total 27,120 gasometer images for training, validation and testing of 3 types of deep neural network. 22,985 images were used for training and validation, 4,135 images were used for testing. We randomly splitted 22,985 images with 8:2 ratio for training and validation respectively for each training epoch. 4,135 test image were categorized into 5 types (Normal, noise, reflex, scale and slant). Normal data is clean image data, noise means image with noise signal, relfex means image with light reflection in gasometer region, scale means images with small object size due to long-distance capturing and slant means images which is not horizontally flat. Final character string recognition accuracies for device ID and gas usage amount of normal data are 0.960 and 0.864 respectively.

• Journal of Intelligence and Information Systems
• /
• v.27 no.4
• /
• pp.73-95
• /
• 2021

This study uses Node2vec graph embedding method and Light GBM link prediction to explore undeveloped export candidate countries in Korea's food and beverage industry. Node2vec is the method that improves the limit of the structural equivalence representation of the network, which is known to be relatively weak compared to the existing link prediction method based on the number of common neighbors of the network. Therefore, the method is known to show excellent performance in both community detection and structural equivalence of the network. The vector value obtained by embedding the network in this way operates under the condition of a constant length from an arbitrarily designated starting point node. Therefore, it has the advantage that it is easy to apply the sequence of nodes as an input value to the model for downstream tasks such as Logistic Regression, Support Vector Machine, and Random Forest. Based on these features of the Node2vec graph embedding method, this study applied the above method to the international trade information of the Korean food and beverage industry. Through this, we intend to contribute to creating the effect of extensive margin diversification in Korea in the global value chain relationship of the industry. The optimal predictive model derived from the results of this study recorded a precision of 0.95 and a recall of 0.79, and an F1 score of 0.86, showing excellent performance. This performance was shown to be superior to that of the binary classifier based on Logistic Regression set as the baseline model. In the baseline model, a precision of 0.95 and a recall of 0.73 were recorded, and an F1 score of 0.83 was recorded. In addition, the light GBM-based optimal prediction model derived from this study showed superior performance than the link prediction model of previous studies, which is set as a benchmarking model in this study. The predictive model of the previous study recorded only a recall rate of 0.75, but the proposed model of this study showed better performance which recall rate is 0.79. The difference in the performance of the prediction results between benchmarking model and this study model is due to the model learning strategy. In this study, groups were classified by the trade value scale, and prediction models were trained differently for these groups. Specific methods are (1) a method of randomly masking and learning a model for all trades without setting specific conditions for trade value, (2) arbitrarily masking a part of the trades with an average trade value or higher and using the model method, and (3) a method of arbitrarily masking some of the trades with the top 25% or higher trade value and learning the model. As a result of the experiment, it was confirmed that the performance of the model trained by randomly masking some of the trades with the above-average trade value in this method was the best and appeared stably. It was found that most of the results of potential export candidates for Korea derived through the above model appeared appropriate through additional investigation. Combining the above, this study could suggest the practical utility of the link prediction method applying Node2vec and Light GBM. In addition, useful implications could be derived for weight update strategies that can perform better link prediction while training the model. On the other hand, this study also has policy utility because it is applied to trade transactions that have not been performed much in the research related to link prediction based on graph embedding. The results of this study support a rapid response to changes in the global value chain such as the recent US-China trade conflict or Japan's export regulations, and I think that it has sufficient usefulness as a tool for policy decision-making.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.23 no.2
• /
• pp.51-58
• /
• 2019

Purpose In in-vitro laboratories of nuclear medicine department, when the reagent lot or reagent lot changes Comparability test or parallel test is performed to determine whether the results between lots are reliable. The most commonly used standard domestic laboratories is to obtain %difference from the difference in results between two lots of reagents, and then many laboratories are set the standard to less than 20% at low concentrations and less than 10% at medium and high concentrations. If the range is deviated from the standard, the test is considered failed and it is repeated until the result falls within the standard range. In this study, several tests are selected that are performed in nuclear medicine in-vitro laboratories to analyze parallel test results and to establish criteria for customized percent difference for each test. Materials and Methods From January to November 2018, the result of parallel test for reagent lot change is analyzed for 7 items including thyroid-stimulating hormone (TSH), free thyroxine (FT4), carcinoembryonic antigen (CEA), CA-125, prostate-specific antigen (PSA), HBs-Ab and Insulin. The RIA-MAT 280 system which adopted the principle of IRMA is used for TSH, FT4, CEA, CA-125 and PSA. TECAN automated dispensing equipment and GAMMA-10 is used to measure insulin test. For the test of HBs-Ab, HAMILTON automated dispensing equipment and Cobra Gamma ray measuring instrument are used. Separate reagent, customized calibrator and quality control materials are used in this experiment. Results 1. TSH [%diffrence Max / Mean / Median] (P-value by t-test > 0.05) C-1(low concentration) [14.8 / 4.4 / 3.7 / 0.0 ] C-2(middle concentration) [10.1 / 4.2 / 3.7 / 0.0] 2. FT4 [%diffrence Max / Mean / Median] (P-value by t-test > 0.05) C-1(low concentration) [10.0 / 4.2 / 3.9 / 0.0] C-2(high concentration) [9.6 / 3.3 / 3.1 / 0.0 ] 3. CA-125 [%diffrence Max / Mean / Median] (P-value by t-test > 0.05) C-1(middle concentration) [9.6 / 4.3 / 4.3 / 0.3] C-2(high concentration) [6.5 / 3.5 / 4.3 / 0.4] 4. CEA [%diffrence Max / Mean / median] (P-value by t-test > 0.05) C-1(low concentration) [9.8 / 4.2 / 3.0 / 0.0] C-2(middle concentration) [8.7 / 3.7 / 2.3 / 0.3] 5. PSA [%diffrence Max / Mean / Median] (P-value by t-test > 0.05) C-1(low concentration) [15.4 / 7.6 / 8.2 / 0.0] C-2(middle concentration) [8.8 / 4.5 / 4.8 / 0.9] 6. HBs-Ab [%diffrence Max / Mean / Median] (P-value by t-test > 0.05) C-1(middle concentration) [9.6 / 3.7 / 2.7 / 0.2] C-2(high concentration) [8.9 / 4.1 / 3.6 / 0.3] 7. Insulin [%diffrence Max / Mean / Median] (P-value by t-test > 0.05) C-1(middle concentration) [8.7 / 3.1 / 2.4 / 0.9] C-2(high concentration) [8.3 / 3.2 / 1.5 / 0.1] In some low concentration measurements, the percent difference is found above 10 to nearly 15 percent in result of target value calculated at a lower concentration. In addition, when the value is measured after Standard level 6, which is the highest value of reagents in the dispensing sequence, the result would have been affected by a hook effect. Overall, there was no significant difference in lot change of quality control material (p-value>0.05). Conclusion Variations between reagent lots are not large in immunoradiometric assays. It is likely that this is due to the selection of items that have relatively high detection rate in the immunoradiometric method and several remeasurements. In most test results, the difference was less than 10 percent, which was within the standard range. TSH control level 1 and PSA control level 1, which have low concentration target value, exceeded 10 percent more than twice, but it did not result in a value that was near 20 percent. As a result, it is required to perform a longer period of observation for more homogenized average results and to obtain laboratory-specific acceptance criteria for each item. Also, it is advised to study observations considering various variables.

• Journal of Life Science
• /
• v.25 no.8
• /
• pp.861-869
• /
• 2015

One of the domesticated species; the dog has been selectively bred for various aims by human. The dog has many breeds, which are artificially selected for specific behaviors and morphologies. Dogs contribute their life to human as working dogs for guide, rescue, detection or etc. Working dogs requires good personality, such as gentleness, robustness and patience for performing their special duty. Many studies have concentrated on finding genetic marker for selecting the high-quality working dog. In this study, we confirmed quantitative expression patterns of eight genes (ABAT; 4-Aminobutyrate Aminotransferase, PLCB1; Phospholipase C, Beta 1, SLC10A4; Solute Carrier Family 10, Member 4, WNT1; Wingless-Type MMTV Integration Site Family, Member 1, BARX2; BarH-Like Homeobox 2, NEUROD6; Neuronal Differentiation 6, SEPT9; Septin 9 and TBR1; T-Box, Brain, 1) among brains tissues from four dog breeds (Beagle, Sapsaree, Shepherd and Jindo), because these genes were expressed and have functions in brain mostly. Specially, BARX2, SEPT9, SLC10A4, TBR1 and WNT1 genes were highly expressed in Beagle and Jindo, and Sapsaree and German Shepherd were vice versa. The biological significance of total genes was estimated by database for annotation, visualization and integrated discovery (DAVID) to determine a different gene ontology (GO) class. In these analyses, we suppose to these eight genes could provide influential information for brain development, and intelligence of organisms. Taken together, these results could provide clues to discover biomarker related to functional traits in brain, and beneficial for selecting superior working dogs.

• Journal of Life Science
• /
• v.28 no.7
• /
• pp.835-840
• /
• 2018

Quantitative determination of the protein expression levels is one of the most important parts in assessment of the safety of foods derived from genetically modified (GM) crops. Overexpression of AtCYP78A7, a gene encoding cytochrome P450 protein, has been reported to improve tolerance to abiotic stress, such as drought and salt stress, in transgenic rice (Oryza sativa L.). In the present study, an enzyme-linked immunosorbent assay (ELISA) kit for diagnosing AtCYP78A7 protein including AtCYP78A7-specific monoclonal antibody was developed. GST-AtCYP78A7 recombinant protein was induced and purified by affinity column. Four monoclonal antibodies (mAb 6A7, mAb 4C2, mAb 11H6, and mAb 7E8) against recombinant protein were also produced and biotinylated with avidin-HRP. After pairing test using GST-AtCYP78A7 protein and lysate of rice samples, mAb 4C2 and mAb 7E8 were selected as a capture antibody and a detecting antibody, respectively, for ELISA kit. Product test using rice samples indicated that percentages of detected protein in total protein were greater than 0.1% in AtCYP78A7-overexpressing transgenic rice (Line 10B-5 and 18A-4), whereas those in negative control non-transgenic rice (Ilpum and Hwayoung) were less than 0.1%. The ELISA kit developed in this study can be useful for the rapid detection and safety assessment of transgenic rice overexpressing AtCYP78A7.

• Analytical Science and Technology
• /
• v.28 no.6
• /
• pp.385-397
• /
• 2015

Harmful and potentially harmful constituents (HPHCs) are chemical compounds in tobacco smoke that cause harm to smokers and non-smokers. This study established and validated methods for the analysis of HPHCs from mainstream cigarette smoke. The analyzed HPHCs were categorized into seven groups: aromatic amines, volatile organic compounds (VOCs), heavy metals, tobacco specific nitrosamines (TSNAs), benzo[a]pyrene (B[a]P), ammonia, and carbonyl compounds. The methods were validated by specificity, linearity, limit of detection (LOD), accuracy, precision, and recovery. These validated methods were then applied to the reference cigarettes (1R5F, 3R4F). The correlation coefficients (r²) for the calibration curves of the seven groups were over 0.995. The LODs showed values of 0.01-0.04 ng/cig cig for aromatic amines, 0.01-0.16 μg/cig for VOCs, 0.01-1.27 ng/cig for heavy metals, 0.06-0.28 ng/cig for TSNAs, 0.04 ng/cig for benzo[a]pyrene, 0.08 μg/cig for ammonia, and 0.78-1.77 μg/cig for carbonyl compounds. The precisions obtained from the intra and inter-day batches were less than 15%. The accuracy and the recovery range were less than 15% and 79.2-117.5%, respectively. The proposed methods can therefore be applied for determining HPHCs in tobacco mainstream smoke.

• Journal of Life Science
• /
• v.19 no.2
• /
• pp.277-283
• /
• 2009

Biogenic amines are generally formed through the decarboxylation of specific free amino acids by exogenous decarboxylases released by microbial species associated with the fish products and fermented feeds. This study was conducted to investigate the properties of e tuna waste regarding the control of degradation of biogenic amines (histamine, tyramine, tryptamine, putrescine, and cadaverine) that might be related with the anti-nutritional factor of the tuna waste that is used for manufacturing domestic fish meal. The values of pH and the salt content were 6.51, 3.35% in tuna waste and 5.58 and 5.83% in tuna fish meal, respectively. The strains and dominant bacteria tested in the tuna waste sample were 9.20, 9.29, 5.67, 7.82 and 7.58 log CFU/g of total bacteria, aerobic plate count (APC), total coliform (TC), Lactobacillus spp. and Bacillus spp., respectively. The main histamine forming-bacteria (HFB) in tuna waste were detected by silica gel thin-layer chromatography (TLC) and 7 histamine-forming bacterial species were isolated among microbes grown in selective medium. The histamine concentration was determined by detection of fluorescence of ο-phthaldialdehyde (OPA) derivatives using HPLC and the date were used to reconfirm the identities of the amine-producing bacteria. The 15 histamine- forming bacteria strains grown in trypicase soy broth (TSB) supplemented with 1% L-histidine (TSBH) were identified as Lactococcus(L.) lactis subsp. lactis, Klebsiella pneummonlae, L. garvieae 36, Vibrio olivaceus, Hafnia alvei and L. garvieae which were main dominant amine - producing strains, and Morganella morganii identified by 16S ribosomal RNA (rRNA) sequencing with PCR amplification. A Phylogenetic tree generated from the 16S rRNA sequencing data showed different phyletic lines that could be readily classified as biogenic amine forming gram-positive and negative bacteria.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.49 no.6
• /
• pp.507-511
• /
• 2004

This research was carried out to offer the basic informations about new varietal breeding for specific use and physiological characteristics through investigation of detection, content variation and color difference of anthocyanin individual pigments within seed coats in domestic black soybean. The seed of thirteen cultivars such as Geomjeongkong 1, Geomjeongkong 2, Seonheugkong, Tawonkong, Ilpumgeomjeongkong, Geomjeongolkong, Cheongjakong, Jinju 1, Heugcheongkong, Juinunikong-Y, Juinunikong-G, Geomjeongkong 3, Geomjeongkong 4 was tested. C3G (cyanidin-3-glucoside) was detected in only Geomjeongkong 1 and Seonheugkong but D3G (delphinidin-3-glucoside) and C3G were found in Heugcheongkong. The rest cultivars that there were three anthocyanins such as D3G, C3G, and Pt3G (petunidin-3-glucoside). Anthocyanin content of tested cultivars showed a high variation. The ranges of D3G, C3G, Pt3G, and TA (total anthocyanin) contents were $0.55\~2.63mg/g,\;2.77\~8.38mg/g,\;0.38\~5.65mg/g,\;and\;3.32\~16.92mg/g$, respectively. These contents showed variation among cultivars as well as variation between two years, 2001-2002. As a result of variation of anthocyanin color difference, the ranges of L (lightness), a (redness), and b (yellowness) as Hunter's value were $34.09\~42.89,\;12.77\!22.85,\;and\;5.36\~12.10$, respectively, and these color differences showed variation among cultivars and also variation between two years, 2001-2002. D3G, C3G, Pt3G, and TA showed reciprocally a positive correlation being representive of high significance.

• Journal of Korean Society of Occupational and Environmental Hygiene
• /
• v.16 no.4
• /
• pp.314-323
• /
• 2006

This study was conducted to identify ingredients of thinners and to confirm reliability of material safety data sheets (MSDS) of thinners for public and workers' health. The 41 thinner products were collected from paint shops located in Busan and Gyeongnam province. The 12 thinner products among them were identified using product MSDS. GC-MSD was used to analyze 41 kinds of thinners qualitatively and quantitatively. The 12 products MSDS were compared with thinner's component through qualitative analysis to confirm MSDS. Chemical ingredients, such as Benzene, Toluene, and Xylene etc., of thinners were analysed in quantity. The 41 thinner products contained 17 disclosed specific, trade name, or generically described chemical solvent ingredients. These 17 ingredients came under 6 classes: alcohols, aromatic hydrocarbons, esters, glycol ethers, ketones, and mixtures. These 17 ingredients were important in the view of industrial hygiene and had occupational exposure limit in the ambient, such as toluene, xylene, acetone, nonane, EGEE, heptane, cumene, MIBK, indene, tri-methyl benzene, etc, were found in 41 kinds of thinners. Aromatic hydrocarbons were the most identified ingredient in thinners. Especially, the benzene, which induces leukemia, was found in 4 kinds of thinners. The content rates of benzene in thinners were 0.25~1.18%. The benzene in enamel thinner, which were 0.39~0.72%, was highest from chemical classification. The contents of toluene, which was found from 27 kinds of thinners, were 5.35~64.16%, which were highest in sobu thinner as 58.80%. Xylene was found from 22 kinds of thinners and contents of xylene were 4.61~72.42%. Acrylic thinner's contents of xylene were 12.06~51.05%, which was most high. It was found that contents of benzene were increased and frequency of detection was decreased through comparison with other study. The MSDS possession rate of paint shops was low as 29.27%. So it did not provide information with public or workers. Mean of agreement rate between MSDS and components of thinners through qualitative analysis was 42.01% and it has wide range from 8.3% to 75%. There are many deficiencies in MSDS about component of thinners. In some case of sample, expecially, despite containing benzene, information was not written it on MSDS.

• Research in Plant Disease
• /
• v.16 no.3
• /
• pp.238-246
• /
• 2010

In 2006 fall, a preliminary survey of viruses in two important medicinal plants, Cynanchum wilfordii and C. auriculatum, was conducted on the experimental fields at the Agricultural Research and Extension Services of Chungbuk province in Korea. On each experimental fields, percentage of virus infection was ranged from 20 to 80%, and especially an average of disease incidence propagated by roots was twice higher than that by seeds. The various symptoms were observed in Cynanchum spp. plants, such as mosaic, mottle, necrosis, yellowing, chlorotic spot and malformation etc. In electron microscopic examination of crude sap extracts, filamentous rod particles with 390-730 nm were observed in most samples. The virus particles were purified from the leaves of C. wilfordii with typical mosaic symptom, and the viral RNA was extracted from this sample containing 430-845 nm long filamentous rod. To identify the viruses, reverse transcription followed by PCR with random primers was carried out. The putative sequences of P3 and coat protein of potyvirus were obtained. From a BLAST of the two sequences, they showed 26-38% and 62-72% identities to potyviruses, respectively. In SDS-PAGE analysis, the subunit of coat protein was approximately 30.3 kDa, close to the coat protein of potyvirus. In bioassay with 21 species in 7 families, Chenopodium quinoa showed local lesion on inoculated leave and chlorotic spot on upper leave, but the others were not infected. RT-PCR detection using specific primer of C. wilfordii and C. auriculatum samples, all of 24 samples with virus symptom was positive, and five out of seven samples without virus symptom were also positive. On the basis of these data, the virus could be considered as a new member of potyvirus. We suggested that the name of the virus was Keunjorong mosaic virus (KjMV) after the common Korean name of C. wilfordii.