• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.1
• /
• pp.1-7
• /
• 2023

An artificial intelligence (AI) method based on image deep learning is proposed to predict the entire displacement shape of a structure using the feature of partial displacements. The performance of the method was investigated through a structural test of a steel frame. An image-to-image regression (I2IR) training method was developed based on the U-Net layer for image recognition. In the I2IR method, the U-Net is modified to generate images of entire displacement shapes when images of partial displacement shapes of structures are input to the AI network. Furthermore, the training of displacements combined with the location feature was developed so that nodal displacement values with corresponding nodal coordinates could be used in AI training. The proposed training methods can consider correlations between nodal displacements in 3D space, and the accuracy of displacement predictions is improved compared with artificial neural network training methods. Displacements of the steel frame were predicted during the structural tests using the proposed methods and compared with 3D scanning data of displacement shapes. The results show that the proposed AI prediction properly follows the measured displacements using 3D scanning.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.4
• /
• pp.159-172
• /
• 2023

In general, social problem-solving research aims to create important social value by offering meaningful answers to various social pending issues using scientific technologies. Not surprisingly, however, although numerous and extensive research attempts have been made to alleviate the social problems and issues in nation-wide, we still have many important social challenges and works to be done. In order to facilitate the entire process of the social problem-solving research and maximize its efficacy, it is vital to clearly identify and grasp the important and pressing problems to be focused upon. It is understandable for the problem discovery step to be drastically improved if current social issues can be automatically identified from existing R&D resources such as technical reports and articles. This paper introduces a comprehensive dataset which is essential to build a machine learning model for automatically detecting the social problems and solutions in various national research reports. Initially, we collected a total of 700 research reports regarding social problems and issues. Through intensive annotation process, we built totally 24,022 sentences each of which possesses its own category or label closely related to social problem-solving such as problems, purposes, solutions, effects and so on. Furthermore, we implemented four sentence classification models based on various neural language models and conducted a series of performance experiments using our dataset. As a result of the experiment, the model fine-tuned to the KLUE-BERT pre-trained language model showed the best performance with an accuracy of 75.853% and an F1 score of 63.503%.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.3
• /
• pp.115-122
• /
• 2022

With the outbreak of COVID-19 a few years ago, video conferencing and electronic document work have increased, and for this reason, the proportion of computer work among modern people's daily routines is increasing. However, as more and more people work on computers in the wrong posture for a long time, the number of patients with poor eyesight and text neck is increasing. Until recently, many studies have been published to correct posture, but most of them have limitations that users may experience discomfort because they have to correct posture by wearing equipment. A posture correction sensor algorithm is proposed to prevent access to the minimum distance between a computer monitor and a person using an ultrasonic sensor device. At this time, an algorithm for minimizing false alarms among warning alarms that sound at the minimum distance is also proposed. Because the ultrasonic sensor device is used, posture correction can be performed without attaching a device to the body, and the user can relieve discomfort. In addition, experimental results showed that accuracy can be improved by reducing false alarms by removing more than half of the noise generated during distance measurement.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.4
• /
• pp.241-248
• /
• 2022

Shear Reinforcement of Dual Anchorage(SRD) is used to reinforce the safety of reinforced concrete structures at construction sites. Welding is used to make shear reinforcement, and welding plays an important role in determining productivity and competitiveness of products. Therefore, a weld bead detection inspection is required. In this paper, we suggest an algorithm for inspecting welding beads using image data of welding beads. First, the proposed algorithm calculates a brightness value in a vertical direction in an image, and then divides a welding bead in a vertical direction by finding a position corresponding to a 50% height point of the brightness value distribution in the image. The welding bead area is also divided in the same way for the horizontal direction, and then the segmentation image is analyzed if there is a welding bead. The proposed algorithm reduced the amount of computation by performing analysis after specifying the region of interest. In addition, accuracy could be improved by using all brightness values in the vertical and horizontal directions using the difference of brightness between the base metal and the welding bead region in the SRD image. The experiment compared the analysis results using five algorithms, such as K-mean and K-neighborhood, as a method to detect if there is a welding bead, and the experimental result proved that the proposed algorithm was the most accurate.

• Animal Bioscience
• /
• v.36 no.4
• /
• pp.654-670
• /
• 2023

Objective: This study aimed to develop loop-mediated isothermal amplification (LAMP) combined with lateral flow dipstick (LFD) and compare it with LAMP-AGE, polymerase chain reaction (PCR), and standard Salmonella culture as reference methods for detecting Salmonella contamination in animal products and animal production environmental samples. Methods: The SalInvA01 primer, derived from the InvA gene and designed as a new probe for LFD detection, was used in developing this study. Adjusting for optimal conditions by temperature, time, and reagent concentration includes evaluating the specificity and limit of detection. The sampling of 120 animal product samples and 350 animal production environmental samples was determined by LAMP-LFD, comparing LAMP-AGE, PCR, and the culture method. Results: Salmonella was amplified using optimal conditions for the LAMP reaction and a DNA probe for LFD at 63℃ for 60 minutes. The specificity test revealed no cross-reactivity with other microorganisms. The limit of detection of LAMP-LFD in pure culture was 3×10² CFU/mL (6 CFU/reaction) and 9.01 pg/μL in genomic DNA. The limit of detection of the LAMP-LFD using artificially inoculated in minced chicken samples with 5 hours of pre-enrichment was 3.4×10⁴ CFU/mL (680 CFU/reaction). For 120 animal product samples, Salmonella was detected by the culture method, LAMP-LFD, LAMP-AGE, and PCR in 10/120 (8.3%). In three hundred fifty animal production environmental samples, Salmonella was detected in 91/350 (26%) by the culture method, equivalent to the detection rates of LAMP-LFD and LAMP-AGE, while PCR achieved 86/350 (24.6%). When comparing sensitivity, specificity, positive predictive value, and accuracy, LAMP-LFD showed the best results at 100%, 95.7%, 86.3%, and 96.6%, respectively. For Kappa index of LAMP-LFD, indicated nearly perfect agreement with culture method. Conclusion: The LAMP-LFD Salmonella detection, which used InvA gene, was highly specific, sensitive, and convenient for identifying Salmonella. Furthermore, this method could be used for Salmonella monitoring and primary screening in animal products and animal production environmental samples.

• Particle and aerosol research
• /
• v.19 no.1
• /
• pp.13-20
• /
• 2023

The Asian dust observation network operates β-ray attenuation samplers to measure PM₁₀ concentrations. In addition, equivalence evaluation and accuracy inspection(Precision Tests) are conducted every year for the reliability of data. β-ray attenuation samplers(16 units) were comparatively observed from May to June 2020 and from July to December 2021. During the observation period, the average daily temperature was the lowest at 6.4℃ in December and the highest at 27.3℃ in August. The average daily humidity ranged from 60% to 100%, but the average daily humidity was over 75% from July to September. The minimum value of the PM₁₀ Gravimetric method was 5.0 ㎍/m³, the maximum value was 53.4 ㎍/m³, and the average value was 17.8 ㎍/m³. The equivalence evaluation results of the PM₁₀ Gravimetric method and β-ray attenuation samplers satisfied the criteria (slope: 1±0.1, intercept: 0±0.5). A relative error analysis between the PM₁₀ Gravimetric method and β-ray attenuation samplers equipment showed that the relative error increased when the concentration was low and the temperature and humidity were high. In addition, in the β-ray attenuation samplers 5-minute interval observation data in May 2020, a relatively large Standard devication was shown as an average maximum ±23.4 ㎍/m³ and a minimum ±15.2 ㎍/m³. At standard deviations of 10% and 90%, equipment with high variability (deviation) was measured at 6 ㎍/m³and 61 ㎍/m³, and equipment with low variability was measured at 12 ㎍/m³ and 47 ㎍/m³. It was confirmed that concentration differences occurred due to differences in variability for each equipment.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.2
• /
• pp.107-114
• /
• 2022

In order to classify 3D film images as right or wrong, it is necessary to detect the pattern in a 3D film image. However, if the contrast of the pixels in the 3D film image is low, it is not easy to classify as the right and wrong 3D film images because the pattern in the image might not be clear. In this paper, we propose a method of classifying 3D film images as right or wrong by comparing the width at a specific frequency of each histogram after obtaining the histogram. Since, it is classified using the width of the histogram, the analysis process is not complicated. From the experiment, the histograms of right and wrong 3D film images were distinctly different, and the proposed algorithm reflects these features, and showed that all 3D film images were accurately classified at a specific frequency of the histogram. The performance of the proposed algorithm was verified to be the best through the comparison test with the other methods such as image subtraction, otsu thresholding, canny edge detection, morphological geodesic active contour, and support vector machines, and it was shown that excellent classification accuracy could be obtained without detecting the patterns in 3D film images.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.2
• /
• pp.102-111
• /
• 2023

Automatic Music Transcription (AMT) is a task that detects and recognizes musical note events from a given audio recording. In this paper, we focus on reducing the latency of real-time AMT systems on piano music. Although neural AMT models have been adapted for real-time piano transcription, they suffer from high latency, which hinders their usefulness in interactive scenarios. To tackle this issue, we explore several techniques for reducing the intrinsic latency of a neural network for piano transcription, including reducing window and hop sizes of Fast Fourier Transformation (FFT), modifying convolutional layer's kernel size, and shifting the label in the time-axis to train the model to predict onset earlier. Our experiments demonstrate that combining these approaches can lower latency while maintaining high transcription accuracy. Specifically, our modified model achieved note F1 scores of 92.67 % and 90.51 % with latencies of 96 ms and 64 ms, respectively, compared to the baseline model's note F1 score of 93.43 % with a latency of 160 ms. This methodology has potential for training AMT models for various interactive scenarios, including providing real-time feedback for piano education.

• Journal of the Korea Society for Simulation
• /
• v.32 no.1
• /
• pp.35-44
• /
• 2023

In recent, there are many research cases that analyze trends or research trends with text analysis. When collecting documents by searching for keywords in abbreviations for data analysis, it is necessary to disambiguate abbreviations. In many studies, documents are classified by hand-work reading the data one by one to find the data necessary for the study. Most of the studies to disambiguate abbreviations are studies that clarify the meaning of words and use supervised learning. The previous method to disambiguate abbreviation is not suitable for classification studies of documents looking for research data from abbreviation search documents, and related studies are also insufficient. This paper proposes a method of semi-automatically classifying documents collected by abbreviations by going topic modeling with Non-Negative Matrix Factorization, an unsupervised learning method, in the data pre-processing step. To verify the proposed method, papers were collected from academic DB with the abbreviation 'MSA'. The proposed method found 316 papers related to Micro Services Architecture in 1,401 papers. The document classification accuracy of the proposed method was measured at 92.36%. It is expected that the proposed method can reduce the researcher's time and cost due to hand work.

• Journal of Environmental Health Sciences
• /
• v.48 no.6
• /
• pp.298-305
• /
• 2022

Background: The concentration of air pollutants as measured by the Air Quality Monitoring System (AQMS) is not an accurate population exposure level since actual human activities and temporal and spatial variability need to be considered. Therefore, to increase the accuracy of exposure assessment, the population should be considered. However, it is difficult to obtain population data due to limitations such as personal information. Objectives: The existing population defined in this study is the number of people in each region's grid. The purpose is to provide a methodology for evaluating exposure to PM_2.5 through existing population data provided by the National Geographic Information Institute. Methods: The selected study period was from October 26 to October 28, 2021. Using PM_2.5 concentration data measured at the Sensor-based Air Monitoring Station (SAMS) installed in Guro-gu and Wonju-si, the concentration for each grid was estimated by applying inverse distance weights through QGIS version 3.22. Considering the existing population, population-weighted average concentration (PWAC) was calculated and the exposure level of the population was compared by region. Results: The outdoor PM_2.5 concentration as measured through the SAMS was high in Wonju-si on all three days. Wonju-si showed an average 22% higher PWAC than Guro-gu. As a result of comparing the PWAC and outdoor PM_2.5 concentration by region, the PWAC in Guro-gu was 1~2% higher than the observed value, but it was almost the same. Conversely, observations of Wonju-si were 10.1%, 11.3%, and 8.2% higher than PWAC. Conclusions: It is expected that the Geographic Information System (GIS) method and the existing population will be used to evaluate the exposure level of a population with a narrow activity radius in further research. In addition, based on this study, it is judged that research on exposure to environmental pollutants and risk assessment methods should be expanded.