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.

Water utilities are making various efforts to reduce water losses from water networks, and an essential part of them is to recognize the moment when a pipe burst occurs during operation quickly. Several physics-based methods and data-driven analysis are applied using real-time flow and pressure data measured through a SCADA system or smart meters, and methodologies based on machining learning are currently widely studied. Water utilities should apply various approaches together to increase pipe burst detection. The most intuitive and explainable water balance method and its procedure were presented in this study, and the applicability and detection performance were evaluated by applying this approach to water supply pipelines. Based on these results, water utilities can establish a mass balance-based pipe burst detection system, give a guideline for installing new flow meters, and set the detection parameters with expected performance. The performance of the water balance analysis method is affected by the water network operation conditions, the characteristics of the installed flow meter, and event data, so there is a limit to the general use of the results in all sites. Therefore, water utilities should accumulate experience by applying the water balance method in more fields.

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.

Anomaly detection for each industrial machine is recognized as one of the essential techniques for machine condition monitoring and preventive maintenance. Anomaly detection of industrial machinery relies on various diagonal data from equipped sensors, such as temperature, pressure, electric current, vibration, and sound, to name a few. Among these data, sound data are easy to collect in the factory due to the relatively low installation cost of microphones to existing facilities. We develop a real time anomalous sound detection (ASD) system with the use of Autoencoder (AE) models in the industrial environments. The proposed processing pipeline makes use of the audio features extracted from the streaming audio signal captured by a single-channel microphone. The pipeline trains AE model by the collected normal sound. In real factory applications, the reconstruction error generated by the trained AE model with new input sound streaming is calculated to measure the degree of abnormality of the sound event. The sound is identified as anomalous if the reconstruction error exceeds the preset threshold. In our experiment on the CNC milling machining, the proposed system shows 0.9877 area under curve (AUC) score.

Fescues, which are widely cultivated as grasses and forages around the world, are often naturally infected with the endophyte, Epichloë. This fungus, transmitted through seeds, imparts resistance to drying and herbivorous insects in its host without causing any external damage, thereby contributing to the adaptation of the host to the environment and maintaining a symbiosis. However, some endophytes, such as E. coenophialum synthesize ergovaline or lolitrem B, which accumulate in the plant and impart anti-mammalian properties. For example, when livestock consume excessive amounts of grass containing toxic endophytes, problems associated with neuromuscular abnormalities, such as convulsions, paralysis, high fever, decreased milk production, reproductive disorders, and even death, can occur. Therefore, pre-inoculation with non-toxic endogenous fungi or management with endophyte-free grass is important in preventing damage to livestock and producing high-quality forage. To date, the diagnosis of endophytes has been mainly performed by observation under a microscope following staining, or by performing an immune blot assay using a monoclonal antibody. Recently, the polymerase chain reaction (PCR)-based molecular diagnostic method is gaining importance in the fields of agriculture, livestock, and healthcare given the method’s advantages. These include faster results, with greater accuracy and sensitivity than those obtained using conventional diagnostic methods. For the diagnosis of endophytes, the nested PCR method is the only available option developed; however, it is limited by the fact that the level of toxic alkaloid synthesis cannot be estimated. Therefore, in this study, we aimed to develop a triplex real-time PCR diagnostic method that can determine the presence or absence of endophyte infection using DNA extracted from seeds within 1 h, while simultaneously detecting easD and LtmC genes, which are related to toxic alkaloid synthesis. This new method was then also applied to real field samples.

A lot of sensor and control signals is generated by an industrial controller and related internet-of-things in discrete manufacturing system. The acquired signals are such records indicating whether several process operations have been correctly conducted or not in the system, therefore they are usually composed of binary numbers. For example, once a certain sensor turns on, the corresponding value is changed from 0 to 1, and it means the process is finished the previous operation and ready to conduct next operation. If an actuator starts to move, the corresponding value is changed from 0 to 1 and it indicates the corresponding operation is been conducting. Because traditional fault detection approaches are generally conducted with analog sensor signals and the signals show stationary during normal operation states, it is not simple to identify whether the manufacturing process works properly via conventional fault detection methods. However, digital control signals collected from a programmable logic controller continuously vary during normal process operation in order to show inherent sequence information which indicates the conducting operation tasks. Therefore, in this research, it is proposed to a recurrent neural network-based fault detection approach for considering sequential patterns in normal states of the manufacturing process. Using the constructed long short-term memory based fault detection, it is possible to predict the next control signals and detect faulty states by compared the predicted and real control signals in real-time. We validated and verified the proposed fault detection methods using digital control signals which are collected from a laser marking process, and the method provide good detection performance only using binary values.

Staphylococcus aureus와 Bacillus cereus는 식중독을 일으키는 주요한 원인균 중 하나로 각종 식품에서 검출되면서 많은 주의가 요구되고 있다. 식중독 발생의 원인균을 신속하고 정확하게 검출할 수 있는 여러 가지 방법 중 DNA 분석을 기반으로 하는 PCR 검출법이 보편적으로 사용되고 있다. 본 연구에서는 샌드위치와 같은 즉석 섭취 식품에서 식중독을 유발할 수 있는 S. aureus와 B. cereus 의 신속검출을 위해 conventional PCR과 real-time PCR의 특이도 및 민감도를 비교하였다. 그 결과, 검출한계 범위가 배양액에서는 cPCR의 경우 S. aureus (104-106 CFU/mL), B. cereus (103-105 CFU/mL) 이였고, real-time PCR의 경우 S. aureus (103-106 CFU/mL), B. cereus (102-105 CFU/mL) 이였다. 식품에서는 cPCR의 경우 S. aureus (104-106 CFU/mL), B. cereus (103-105 CFU/mL) 이고, real-time PCR의 경우 S. aureus (104-106 CFU/mL), B. cereus (103-105 CFU/mL) 으로 나타났다. Real-time PCR법이 cPCR법 보다 10배 이상 더 민감한 것으로 나타났다. 따라서 real-time PCR은 우수한 민감도를 지닌 검출기법으로 식중독 세균 검출에 있어 매우 효과적인 방법으로 사료된다.

The leading source of occupational fatalities is a portable ladder in Korea because it is widely used in industry as work platform. In order to reduce victims, it is necessary to establish preventive measures for the accidents caused by portable ladder. Therefore, this study statistically analyzed injury death by portable ladder for recent 10 years to investigate the accident characteristics. Next, to monitor wearing of safety helmet in real-time while working on a portable ladder, this study developed an object detection model based on the You Only Look Once(YOLO) architecture, which can accurately detect objects within a reasonable time. The model was trained on 6,023 images with/without ladders and safety helmets. The performance of the proposed detection model was 0.795 for F1 score and 0.843 for mean average precision. In addition, the proposed model processed at least 25 frames per second which make the model suitable for real-time application.

적조가 처음 시작되는 해역을 조기에 파악하기 위하여 Quantitative real-time PCR (qPCR)을 경남해역 적조현장에 활용하였다. 2019년 경남해역을 대상으로 Cochlodinium polykrikoides를 qPCR로 정량분석한 결과, 6월 초에 저밀도로(0.0015~0.0058 cells mL-1) 검출되기 시작하여 8월 중순에는 최대 0.163 cells mL-1 밀도로 증가하였고, 주로 남해도 주변에서 높게 검출되었다. 8월 말에는 현미경 검경으로 남해도 주변에서 높게 출현함이 확인되었고(최대 24 cells mL-1), 9월 2일에는 남해도에서 적조주의보가 발령되었고(최대 200 cells mL-1), 9월 11일에는 최대 12,000 cells mL-1까지 남해도 해역에서 발생하였다. 위 결과는 극미량의 C. polykrikoides이 적조발생 전에 남해도에서 검출 되었고 이후 같은 해역에서 적조가 발생되었음을 보여준다. 이는 qPCR이 극미량의 C. polykrikoides을 조기검출하는데 유용한 방법임을 보여준다.

This study constructed a wheat-specific primer and a probe using the internal transcribed spacer (ITS). 2 regions of wheat (Triticum aestivum) and real-time PCR conditions were established. The calibration curve showed a slope of -3.356, a correlation coefficient of 0.998, and an amplification efficiency of 98.589%. Experiments were carried out on the rice flour mixed with 50%, 10%, 1%, 0.1%, 0.01%, and 0.001% of wheat. The result showed that it was possible to detect samples mixed with up to 0.01% of wheat. As a result of checking, the wheat detection potential of rice, 34 processed foods, and seven processed foods was ascertained. The real-time PCR method using the wheatspecific primer and probe developed in this study can be used to identify the authenticity of the raw materials, such as the incorrect indication of the raw materials utilized and the unintended mixing of wheat during the manufacturing process.

최근 한국에서 발생한 Salmonella로 인한 식중독 사고 는 2018년 9월 학교급식에서 제공된 초콜릿 무스 케이크가 원인이 되었다. 이 연구의 목적은 Salmonella Typhimurium이 인위적으로 접종된 무스케이크와 티라미수에서 3M Molecular Detection Assay 2 –Salmonella와 식품공전에 등재된 방법인 분리배지와 real-time PCR을 비교하는 것이었다. 무스케이크 2종과 티라미수 2종 25 g에 225 mL BPW를 넣고 37oC에서 24시간 동안 증균 배양하 였다. 배양 후, 3M Molecular Detection Assay 2 – Salmonella, 분리배지 그리고 real-time PCR로 분석하였다. 초콜릿 무스 케이크를 제외하고 3가지 방법은 유사한 결 과를 보였다. 초콜릿 무스 케이크에서 분리배지와 3M Molecular Detection Assay 2 –Salmonella는 모든 접종수 준에서 동일한 결과를 나타낸 반면 real-time PCR은 104 CFU/25g수준에서 1번의 양성결과를 제외하고 모두 검출 되지 않았다. 초콜릿 무스에 S. Typhimurium을 102 CFU/ 25 g 수준으로 접종하였을때, real-time PCR를 이용한 검출은 15%에서는 부분적인 음성을 나타냈고, 20-100% 함량의 초콜릿 무스에서는 모두 음성이었다. Real-time PCR 로는 chocolate이 15% 이상 함유된 식품에서의 Salmonella 균 검출이 불가능하였지만, LMAP 기반의 3M Molecular Detection Assay 2으로는 chocolate 농도에 관계없이 검출이 가능하였다.

식품에 존재하는 병원균을 신속검출하기 위한 방법으로 LAMP와 real-time PCR 방법을 비교 평가 하였다. S. Typhimurium, L. monocytogenes, C. sakazakii의 3종에 대해 식품공전에서 권고하는 식품 종류를 선별하여 민감도를 분석하였다. S. Typhimurium에서는 11종의 식품(햄, 닭가슴살, 계란, 돼지고기, 소고기, 오리고기, 액상음료, 샐러드, 콘프레이크, 초콜릿, 사료)중 4종(햄, 돼지고기, 시리얼, 사료)에서 LAMP보다 real-time PCR에서 검출 민감도가 10 배 이상 더 높았고, 6종(닭가슴살, 계란, 소고기, 오리고기, 음료, 샐러드)에서는 real-time PCR과 비슷한 수준을 그리고 초콜릿에서는 real-time PCR로는 검출되지 않았으며 LAMP로만 검출되는 결과가 나타났다. L. monocytogenes 와 C. sakazakii에서는 9종 모두에서 LAMP보다 real-time PCR에서 검출 민감도가 더 높았다. 또한 L. monocytogenes 에서 LAMP의 검출 민감도가 S. Typhimurium과 C. sakazakii 보다 10배 이상 낮았다. 3M MDS의 검출한계 향상을 위해 변형된 3M MDS의 민감도는 기존대비 10배 이상 증가되었다. 따라서 식품에 존재하는 병원균의 검출 을 위해 식품의 구성성분에 따라 LAMP와 real-time PCR 를 적절히 선택하는 것이 바람직할 것으로 생각되었다. 한편, 농축 방법을 이용해 LAMP방법의 민감도를 향상시킬 수 있음을 알 수 있었다.

기후온난화 현상이 지속되고 있는 제주지역에서 환경적으로 다른 지역 모기의 계절적 발생밀도를 조사하기 위해 제주시의 국제공항, 항만 구역과 축사 그리고 서귀포 도심지의 11지점을 선정하여 3월부터 11월까지 매달 2회씩 Black light trap과 BG sentinel trap을 이용하여 모기를 채집하였다. 채집된 모기는 5속 7종. 6,042마리였으며, 이 중 빨간집모기(Culex pipiens)가 4,159마리(68.8%)로 우점종이었으며 흰줄숲모기 (Aedes albopictus)는 1,348마리(24.4%)였다. Black light trap를 이용한 채집에서 중앙동주민센터는 트랩당 72.8마리를 채집하여 모기 밀도가 가장 높게 나타났으며 제주국제공항은 트랩당 1.4마리로 가장 낮게 나타났다. BG sentinel trap을 이용한 채집에서는 항만에서 트랩당 71.7마리로 가장 많았고 도심지의 걸매생태공원에서 28.3마리로 가장 낮았다. 시기별로 모기의 밀도는 5월부터 증가하기 시작하여 8월에 1,156마리 (19.1%)로 가장 높은 밀도를 나타내었다. 채집된 암컷모기를 종별, 시기별, 지점별로 나누어 pool당 50마리 이하로 설정하여 총 364 pools에서 flavivirus 존재여부를 real time RT-PCR로 검사하였으나, 검출되지 않았다.

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.