In recent automated manufacturing systems, compressed air-based pneumatic cylinders have been widely used for basic perpetration including picking up and moving a target object. They are relatively categorized as small machines, but many linear or rotary cylinders play an important role in discrete manufacturing systems. Therefore, sudden operation stop or interruption due to a fault occurrence in pneumatic cylinders leads to a decrease in repair costs and production and even threatens the safety of workers. In this regard, this study proposed a fault detection technique by developing a time-variant deep learning model from multivariate sensor data analysis for estimating a current health state as four levels. In addition, it aims to establish a real-time fault detection system that allows workers to immediately identify and manage the cylinder’s status in either an actual shop floor or a remote management situation. To validate and verify the performance of the proposed system, we collected multivariate sensor signals from a rotary cylinder and it was successful in detecting the health state of the pneumatic cylinder with four severity levels. Furthermore, the optimal sensor location and signal type were analyzed through statistical inferences.

In the realm of dental prosthesis fabrication, obtaining accurate impressions has historically been a challenging and inefficient process, often hindered by hygiene concerns and patient discomfort. Addressing these limitations, Company D recently introduced a cutting-edge solution by harnessing the potential of intraoral scan images to create 3D dental models. However, the complexity of these scan images, encompassing not only teeth and gums but also the palate, tongue, and other structures, posed a new set of challenges. In response, we propose a sophisticated real-time image segmentation algorithm that selectively extracts pertinent data, specifically focusing on teeth and gums, from oral scan images obtained through Company D's oral scanner for 3D model generation. A key challenge we tackled was the detection of the intricate molar regions, common in dental imaging, which we effectively addressed through intelligent data augmentation for enhanced training. By placing significant emphasis on both accuracy and speed, critical factors for real-time intraoral scanning, our proposed algorithm demonstrated exceptional performance, boasting an impressive accuracy rate of 0.91 and an unrivaled FPS of 92.4. Compared to existing algorithms, our solution exhibited superior outcomes when integrated into Company D's oral scanner. This algorithm is scheduled for deployment and commercialization within Company D's intraoral scanner.

On pig farms, the highest mortality rate is observed among nursing piglets. To reduce this mortality rate, farmers need to carefully observe the piglets to prevent accidents such as being crushed and to maintain a proper body temperature. However, observing a large number of pigs individually can be challenging for farmers. Therefore, our aim was to detect the behavior of piglets and sows in real-time using deep learning models, such as YOLOv4-CSP and YOLOv7-E6E, that allow for real-time object detection. YOLOv4-CSP reduces computational cost by partitioning feature maps and utilizing Cross-stage Hierarchy to remove redundant gradient calculation. YOLOv7-E6E analyzes and controls gradient paths such that the weights of each layer learn diverse features. We detected standing, sitting, and lying behaviors in sows and lactating and starving behaviors in piglets, which indicate nursing behavior and movement to colder areas away from the group. We optimized the model parameters for the best object detection and improved reliability by acquiring data through experts. We conducted object detection for the five different behaviors. The YOLOv4-CSP model achieved an accuracy of 0.63 and mAP of 0.662, whereas the YOLOv7-E6E model showed an accuracy of 0.65 and mAP of 0.637. Therefore, based on mAP, which includes both class and localization performance, YOLOv4-CSP showed the superior performance. Such research is anticipated to be effectively utilized for the behavioral analysis of fattening pigs and in preventing piglet crushing in the future.

The entire industry is increasing the use of big data analysis using artificial intelligence technology due to the Fourth Industrial Revolution. The value of big data is increasing, and the same is true of the production technology. However, small and medium -sized manufacturers with small size are difficult to use for work due to lack of data management ability, and it is difficult to enter smart factories. Therefore, to help small and medium -sized manufacturing companies use big data, we will predict the gross production time through machine learning. In previous studies, machine learning was conducted as a time and quantity factor for production, and the excellence of the ExtraTree Algorithm was confirmed by predicting gross product time. In this study, the worker's proficiency factors were added to the time and quantity factors necessary for production, and the prediction rate of LightGBM Algorithm knowing was the highest. The results of the study will help to enhance the company's competitiveness and enhance the competitiveness of the company by identifying the possibility of data utilization of the MES system and supporting systematic production schedule management.

It is important to ensure worker’s safety from radiation hazard in decommissioning site. Real-time tracking of worker’s location is one of the factors necessary to detect radiation hazard in advance. In this study, the integrated algorithm for worker tracking has been developed to ensure the safety of workers. There are three essential techniques needed to track worker’s location, which are object detection, object tracking, and estimating location (stereo vision). Above all, object detection performance is most important factor in this study because the performance of tracking and estimating location is depended on worker detection level. YOLO (You Only Look Once version 5) model capable of real-time object detection was applied for worker detection. Among the various YOLO models, a model specialized for person detection was considered to maximize performance. This model showed good performance for distinguishing and detecting workers in various occlusion situations that are difficult to detect correctly. Deep SORT (Simple Online and Realtime Tracking) algorithm which uses deep learning technique has been considered for object tracking. Deep SORT is an algorithm that supplements the existing SORT method by utilizing the appearance information based on deep learning. It showed good tracking performance in the various occlusion situations. The last step is to estimate worker’s location (x-y-z coordinates). The stereo vision technique has been considered to estimate location. It predicts xyz location using two images obtained from stereo camera like human eyes. Two images are obtained from stereo camera and these images are rectified based on camera calibration information in the integrated algorithm. And then workers are detected from the two rectified images and the Deep SORT tracks workers based on worker’s position and appearance between previous frames and current frames. Two points of workers having same ID in two rectified images give xzy information by calculating depth estimation of stereo vision. The integrated algorithm developed in this study showed sufficient possibility to track workers in real time. It also showed fast speed to enable real-time application, showing about 0.08 sec per two frames to detect workers on a laptop with high-performance GPU (RTX 3080 laptop version). Therefore, it is expected that this algorithm can be sufficiently used to track workers in real decommissioning site by performing additional parameter optimization.

본 연구는 실시간 원격수업에서 학생들의 학습 장애 요인을 분석하고, 그러한 장애 요인을 해소 할 수 있는 수업을 설계하였으며, 이를 실제로 적용한 사례 연구이다. 2015 개정 교육과정에서 강조 하고 있는 ‘역량’을 함양할 수 있는 실시간 원격수업을 계획하였다. 이 수업은 개별 한자의 모양·음· 뜻과 필순에 대한 지식 암기가 아니라, 한자 정보를 찾아 활용할 수 있는 능력을 함양하는 데에 있다고 할 수 있다. 이를 위해 학생들이 실시간 원격수업에서 겪은 학습 장애 요인을 조사하였다. 응답 내용은 ‘집중’, ‘기술’, ‘학습’, ‘소통’, ‘피로’, ‘흥미’, ‘기타’ 등으로 분류할 수 있었다. 이러한 학습 장애 요인에 대한 개선 방안은 수업의 내적 요인과 수업의 외적 요인으로 나누어 살펴볼 수 있다. 수업의 내적 요인으로는 ‘학습, 흥미’의 문제, 수업의 외적 요인은 ‘집중, 기술, 소통, 피로’의 문제이다. 학습 부담이나 학습 부진에 관한 문제는 화상회의 플랫폼을 통하여 실시간 쌍방향 원격수업 실시, 학습량 적정화 등으로 개선 방안을 모색하고, 흥미는 대면 수업에서 친구들 과 함께 수업 듣는 분위기 조성 등으로 개선 방안을 찾는다. 집중은 활동 중심의 실시간 원격수업, 기술은 학습 플랫폼 오류에 대비한 대체 수단 마련, 소통은 실시간 쌍방향 수업과 피드백 강화, 피로는 일과 시간의 변경 등을 통해 개선 방안을 찾는다. 이러한 응답 내용을 토대로 학생들이 집중할 수 있는 활동 중심 한문 수업, 수업 시간 동안 해결 할 수 있는 적절한 학습량의 과제 제시를 통해 학습 부담 감소, 교사와 소통할 수 있는 채널 확보, 흥미를 가질 수 있는 요소를 가미한 수업 방안을 강구하였다. 기술적인 문제나 피로는 개별 교과에서 해결할 수 없는 문제이기 때문에 수업 구성의 고려 사항에서 제외하였다. 중학교 1학년의 학년 초에 실시한 수업이기 때문에 한자 영역을 대상으로 하였다. 수업은 ‘학습지 제공 → 신습 한자 선택 → 인터넷 사전 검색 → 필순에 따라 한자 쓰기 → 한자를 선택한 이유 쓰기’ 순서로 진행하였다. 처음 한자를 써보는 학생들이 다수였지만, 대부분 정해진 시간 안에 한자의 모양을 정확하게 작성하 였다. 일부 학생은 정해진 수업 시간 동안 학습지를 완성하지 못하였다. 원격수업에서 수업 환경은 교실에서 대면으로 이루어졌던 학교 수업을 그대로 온라인으로 옮겨가는 것이 아니라, 본질적인 수업 방법의 변화가 있어야 할 것이다.

The sensory stimulation of a cosmetic product has been deemed to be an ancillary aspect until a decade ago. That point of view has drastically changed on different levels in just a decade. Nowadays cosmetic formulators should unavoidably meet the needs of consumers who want sensory satisfaction, although they do not have much time for new product development. The selection of new products from candidate products largely depend on the panel of human sensory experts. As new product development cycle time decreases, the formulators wanted to find systematic tools that are required to filter candidate products into a short list. Traditional statistical analysis on most physical property tests for the products including tribology tests and rheology tests, do not give any sound foundation for filtering candidate products. In this paper, we suggest a deep learning-based analysis method to identify hand cream products by raw electric signals from tribological sliding test. We compare the result of the deep learning-based method using raw data as input with the results of several machine learning-based analysis methods using manually extracted features as input. Among them, ResNet that is a deep learning model proved to be the best method to identify hand cream used in the test. According to our search in the scientific reported papers, this is the first attempt for predicting test cosmetic product with only raw time-series friction data without any manual feature extraction. Automatic product identification capability without manually extracted features can be used to narrow down the list of the newly developed candidate products.

For the efficient teaching and learning of Vietnamese, the researcher paid attention to Project Based Learning and tried to apply it to the class. The researcher analyzed prior studies of PBL classes, including English and other foreign languages, and applied the theory of PBL to Vietnamese language education subjects, designed PBL classes, and utilized them in classes. In addition, the process in which the learners perform tasks (student presentation, peer-faculty evaluation, evaluation opinion reflection process), the results, and the questionnaire survey on learners were analyzed. As a result, it was found that PBL methods could also be applied in Vietnamese classes. The learners reorganized the learning contents into his or her own knowledge in the relationship between the learner’s own thoughts, experiences, knowledge, and understanding by referring to the instructor’s teaching plan and lecture. In addition, it was possible to achieve more useful and viable knowledge by listening to other people’s opinions and thoughts about their own knowledge, understanding, and interpretation, and through correction and supplementation processes. Also noteworthy is that through the PBL class, the level of knowledge of each student increased rapidly.

Non-face-to-face online education which suddenly began with COVID-19, was an opportunity to expand to university education sites. This study presents online classes based on the Tandem learning method as examples and examines the class operation, learning content, and learner satisfaction according to the actual type of flip learning classes and real-time video classes. By converting Tandem classes operated by B University in Busan into online classes, the existing e-Tandem classes were expanded using flip learning and ZOOM. The first Flip learning class allows self-directed learning, and the free form of learning through pre-learning becomes an advantage of the online class type. However, while professors participate in the small conference room and feedback, there is a need to be supplemented with the ability to feedback directly to other teams. Furthermore, as a result of the learner satisfaction survey, there were complaints about prior learning and the amount of tasks, so studies on specific tasks and the content and methods of prior learning are also needed. The second real-time video class allows interaction between professors and learners, learners and learners. The biggest feature of this class type is that it can solve the absence of communication, which was a disadvantage of non-face-to-face classes. However, the ability of a professor is needed to conduct a real-time video class like this. Unlike learners who are familiar with digital technology, only when they understand and learn various online content and functions will their online classes become as natural as face-to-face classes.

This paper presents a real-time, false-pick filter based on deep learning to reduce false alarms of an onsite Earthquake Early Warning (EEW) system. Most onsite EEW systems use P-wave to predict S-wave. Therefore, it is essential to properly distinguish P-waves from noises or other seismic phases to avoid false alarms. To reduce false-picks causing false alarms, this study made the EEWNet Part 1 'False-Pick Filter' model based on Convolutional Neural Network (CNN). Specifically, it modified the Pick_FP (Lomax et al.) to generate input data such as the amplitude, velocity, and displacement of three components from 2 seconds ahead and 2 seconds after the P-wave arrival following one-second time steps. This model extracts log-mel power spectrum features from this input data, then classifies P-waves and others using these features. The dataset consisted of 3,189,583 samples: 81,394 samples from event data (727 events in the Korean Peninsula, 103 teleseismic events, and 1,734 events in Taiwan) and 3,108,189 samples from continuous data (recorded by seismic stations in South Korea for 27 months from 2018 to 2020). This model was trained with 1,826,357 samples through balancing, then tested on continuous data samples of the year 2019, filtering more than 99% of strong false-picks that could trigger false alarms. This model was developed as a module for USGS Earthworm and is written in C language to operate with minimal computing resources.

Anomaly detection of Machine Learning such as PCA anomaly detection and CNN image classification has been focused on cross-sectional data. In this paper, two approaches has been suggested to apply ML techniques for identifying the failure time of big time series data. PCA anomaly detection to identify time rows as normal or abnormal was suggested by converting subjects identification problem to time domain. CNN image classification was suggested to identify the failure time by re-structuring of time series data, which computed the correlation matrix of one minute data and converted to tiff image format. Also, LASSO, one of feature selection methods, was applied to select the most affecting variables which could identify the failure status. For the empirical study, time series data was collected in seconds from a power generator of 214 components for 25 minutes including 20 minutes before the failure time. The failure time was predicted and detected 9 minutes 17 seconds before the failure time by PCA anomaly detection, but was not detected by the combination of LASSO and PCA because the target variable was binary variable which was assigned on the base of the failure time. CNN image classification with the train data of 10 normal status image and 5 failure status images detected just one minute before.

Three CNN (Convolutional Neural Network) models of GoogLeNet, VGGNet, and Alexnet were evaluated to select the best deep learning based image analysis mothod that can detect pavement distresses of pothole, spalling, and punchout on expressway. Education data was obtained using pavement surface images of 11,056km length taken by Gopro camera equipped with an expressway patrol car. Also, deep learning framework of Caffe developed by Berkeley Vision and Learning Center was evaluated to use the three CNN models with other frameworks of Tensorflow developed by Google, and CNTK developed by Microsoft. After determing the optimal CNN model applicable for the distress detection, the analyzed images and corresponding GPS locations, distress sizes (greater than distress length of 150mm), required repair material quantities are trasmitted to local maintenance office using LTE wireless communication system through ICT center in Korea Expressway Corporation. It was found out that the GoogLeNet, AlexNet, and VGG-16 models coupled with the Caffe framework can detect pavement distresses by accuracy of 93%, 86%, and 72%, respectively. In addition to four distress image groups of cracking, spalling, pothole, and punchout, 22 different image groups of lane marking, grooving, patching area, joint, and so on were finally classified to improve the distress detection rate.

Background: Deficiencies in the ability to maintain balance are common in elderly. Augmented feedback such as knowledge of results (KR) can accelerate learning and mastering a motor skill in older people. Objects: We designed this study to examine whether one session of Wii-Fit game with self-regulated KR is effective for elderly people, and to compare the effect of two different timings of self-regulated KR conditions. Methods: Thirty-nine community-dwelling elders, not living in hospice care or a nursing home, participated in this study. During acquisition, two groups of volunteers were trained in 10 blocks of a dynamic balancing task under the following 2 conditions, respectively: (a) a pre-trial self-regulated KR (n1=18), or (b) a post-trial self-regulated KR (n2=21). Immediate retention tests and delayed retention tests of balancing performance were administered in 15 minutes and 24 hours following acquisition period, respectively. Results: In both groups, significant improvements of balancing performances scores were observed during the acquisition period. Regardless of the group, mean of balancing performance scores on retention tests were well-maintained from the final session. There were no significant differences between groups in balancing performance scores during the acquisition period (p>.05); however, the post-trial self-regulated KR group exhibited significantly higher balancing performance scores in both the immediate retention test and delayed retention test than that of the pre-trial self-regulated KR group (p<.05). Conclusion: Therefore, subjects who regulated their feedback after a dynamic balancing task, during the acquisition period, experienced more efficient motor learning during the retention period than did subjects who regulated their feedback before a dynamic balancing task. Accordingly, in case of presenting the KR of motor learning in clinical settings to elders who reduced dynamic balance abilities, the requesting time of KR is imperative according to self-estimation processes as well as types of KR and practice.

목적 : 운동학습은 신체적인 연습과 경험의 결과로 동작에 대한 기술 습득의 과정으로 표현된다. 최근 정신적 훈련을 통해 운동 기술의 습득이 이루어질 수 있다는 연구가 많이 보고되고 있다. 본 연구는 시열반응과제를 통하여 실제 훈련과 정신 훈련의 효과에 대해 연구하고자 하였다. 연구방법 : 50명의 정상 성인을 대상으로 정신 훈련군, 실제 훈련군, 대조군으로 무선적 배치하였다. 측정된 시열반응과제는 컴퓨터 화면에 제시되는 네 가지의 색깔에 대응하는 반응패드를 손가락을 이용하여 최대한 빨리 정확히 누르도록 하고, 반응 시간을 측정하였다. 정신 훈련군과 실제 훈련군은 정신 및 실제 훈련을 하루 30분, 2주 동안 훈련하였고, 훈련 시간은 동일하게 설계하였다. 결과 : 이요인 반복 측정 공분산 분석의 집단 간의 주 효과 검정에서 통계학적으로 유의한 차이가 관찰되었으며, 집단×훈련 전·후 간의 상호작용에서도 유의한 차이가 관찰되었다(p<.05). 또한 사후 검정에서 정신 훈련군과 대조군, 실제 훈련군과 대조군에서 유의한 차이를 보였다(p<.05). 이러한 결과는 훈련 후 대조군에서의 시간의 감소보다 정신 훈련군과 실제 훈련군에서의 과제 수행 시 시간의 감소가 더 크게 나타났다. 결론 : 본 연구에서는 정신 훈련군에서 실제 훈련군과 같이 운동 학습의 효과가 나타났으며, 정신 훈련은 효과적인 운동학습을 위한 방법임이 입증되었다. 그러므로 운동제어와 인지학습을 시켜야 하는 재활 영역에서 뇌손상 환자에게 학습을 시키기 위한 방법으로 정신 훈련을 채택할 수 있으며, 마비측의 실제 훈련을 수행하기에 어려운 상황에서는 정신 훈련이 더 효과적일 수 있고, 실제 훈련과 비슷한 효과를 기대할 수 있을 것으로 생각된다.