• 국제학술발표논문집
• /
• The 6th International Conference on Construction Engineering and Project Management
• /
• pp.179-182
• /
• 2015

Detecting and assessing human-related risks is critical to improve the on-site safety condition and reduce the loss in lives, time and budget for construction industry. Recent research in neural science and psychology suggest inattentional blindness that caused by overload in working memory is the major cause of unexpected human related accidents. Due to the limitation of human mental workload, laborers are vulnerable to unexpected hazards while focusing on complicated and dangerous construction tasks. Therefore, detecting the risk perception abilities of workers could help to identify vulnerable individuals and reduce unexpected injuries. However, there are no available measurement approaches or devices capable of monitoring construction workers' mental conditions. The research proposed in this paper aims to develop such a measurement framework to evaluate hazards through monitoring electroencephalogram of labors. The research team developed a wearable safety monitoring helmet, which can collect the brain waves of users for analysis. A bispectrum approach has been developed in this paper to enrich the data source and improve accuracy.

• 한국신재생에너지학회:학술대회논문집
• /
• 한국신재생에너지학회 2007년도 추계학술대회 논문집
• /
• pp.174-176
• /
• 2007

Because of accident or leak of electricity, high voltage electricity can be conducted to vehicle chassis and damage human. Therefore the unit for detecting ground fault is necessary to minimize loss of life or equipment damage. Isolation resistance must be monitored for detecting ground fault. GFD(Ground Fault Detection) unit continually generate the pulse voltage between high voltage network and chassis. This will be sensing the returned current, calculate the isolation resistance and make decision the ground fault. This paper describes the method detecting ground fault.

• 전기전자학회논문지
• /
• 제24권3호
• /
• pp.783-790
• /
• 2020

본 논문에서는 실내 보안 응용을 위한 사람 감지 레이다 시스템을 제안하고, 이의 FPGA 기반 설계 및 구현 결과를 제시하였다. 연산의 복잡도와 메모리 요구량을 최소화하기 위해 스펙트로그램의 상측 절반만 특징점 추출에 사용하였으며, 복잡한 연산이 필요한 특징점 추출기법을 배제하고, 분류 성능과 연산 복잡도를 고려한 효율적인 특징점 추출기법이 제안되었다. 또한, 전체 스펙트로그램에 대한 저장이 불필요한 파이프라인 구조로 설계하여 메모리 요구량을 최소화하였다. 제안된 시스템의 분류 학습을 위해 사람, 개, 로봇 청소기에 대한 실험이 수행되었고, 96.2%의 정확도 성능을 확인하였다. 제안된 시스템은 Verilog-HDL을 이용하여 구현되었으며, 1140개의 logic과 6.5 Kb의 메모리를 사용하는 저면적 설계가 가능함을 확인하였다.

• International Journal of Internet, Broadcasting and Communication
• /
• 제12권4호
• /
• pp.83-92
• /
• 2020

A recent study to determine the fall is focused on analyzing fall motions using a recurrent neural network (RNN), and uses a deep learning approach to get good results for detecting human poses in 2D from a mono color image. In this paper, we investigated the improved detection method to estimate the position of the head and shoulder key points and the acceleration of position change using the skeletal key points information extracted using PoseNet from the image obtained from the 2D RGB low-cost camera, and to increase the accuracy of the fall judgment. In particular, we propose a fall detection method based on the characteristics of post-fall posture in the fall motion analysis method and on the velocity of human body skeleton key points change as well as the ratio change of body bounding box's width and height. The public data set was used to extract human skeletal features and to train deep learning, GRU, and as a result of an experiment to find a feature extraction method that can achieve high classification accuracy, the proposed method showed a 99.8% success rate in detecting falls more effectively than the conventional primitive skeletal data use method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권3호
• /
• pp.1189-1204
• /
• 2018

Detecting and capturing 3D human structures from the intensity-based image sequences is an inherently arguable problem, which attracted attention of several researchers especially in real-time activity recognition (Real-AR). These Real-AR systems have been significantly enhanced by using depth intensity sensors that gives maximum information, in spite of the fact that conventional Real-AR systems are using RGB video sensors. This study proposed a depth-based routine-logging Real-AR system to identify the daily human activity routines and to make these surroundings an intelligent living space. Our real-time routine-logging Real-AR system is categorized into two categories. The data collection with the use of a depth camera, feature extraction based on joint information and training/recognition of each activity. In-addition, the recognition mechanism locates, and pinpoints the learned activities and induces routine-logs. The evaluation applied on the depth datasets (self-annotated and MSRAction3D datasets) demonstrated that proposed system can achieve better recognition rates and robust as compare to state-of-the-art methods. Our Real-AR should be feasibly accessible and permanently used in behavior monitoring applications, humanoid-robot systems and e-medical therapy systems.

• 스마트미디어저널
• /
• 제5권3호
• /
• pp.49-53
• /
• 2016

인체 전자기장 신호를 검출하여 동작 인식에 이용하는 비접촉방식의 방법은 시간과 공간의 제약이 기존의 시스템보다 덜하므로 관련 연구들이 진행 중에 있다. 본 논문에서는 비접촉방식의 인체전기장 신호를 검출할 수 있는 하드웨어를 구현하여 이를 디지털 파형화 하여 인식률을 높일 수 있는 회로시스템을 설계하였다. 차동 증폭회로의 구현과 비교기를 연동한 디지털 파형화를 위한 회로 시스템을 시뮬레이션과 결합하여 PCB화한 후/ 설계된 전체 회로 시스템에 대한 특성평가를 수행하였다.

• 센서학회지
• /
• 제7권3호
• /
• pp.197-202
• /
• 1998

맥파신호는 심장과 심혈관계에 관련된 중요한 정보를 포함하고 있는 중요한 생체신호이다. 본 연구에서는 맥파신호를 비관혈적으로 검출하기 위해 기존의 기계적 변환방식의 개념에 근거하여 광섬유를 이용하는 맥파 검출용 트랜스듀서를 새로이 개발하여 사용하였다. 기계적 변환방식에서 발생될 수 있는 맥동전달 효율을 개선시키기 위해 본 연구에서는 맥동하는 피부와 접촉하여 센서에 전달하는 검출부(detecting part)와, 전달받은 변위운동을 광 출력 변화에 따른 전기 신호로 변환하는 감지부(sensing part)의 두 부분이 조합된 구조를 갖는 설계방식을 제안하였다. 우리는 이러한 조합형 구조의 설계방식을 이용한 결과로 기존의 트랜스듀서로 검출이 어려운 맥파의 C점(절흔)과 T파를 검출할 수 있음을 확인하였다.

• BMB Reports
• /
• 제29권3호
• /
• pp.192-199
• /
• 1996

A rapid and sensitive method has been developed to detect a-fetoprotein (AFP) in human serum by a two-site sandwich enzyme-linked immunosorbent assay (ELISA) with monoclonal antibodies (MAbs) for human AFP within 1 h. To obtain the most sensitive and reliable MAbs. 12 kinds of MAbs (HPJ1 to HPJ12) as a capture antibody and 4 kinds of horseradish peroxidase (HRP) conjugated MAbs as a tracer antibody were investigated. Among these, only HPJ 10-HRP conjugated HPJ 1 (HPJ 10-HPJ $1^*$) and HPJ 11-HRP conjugated HPJ 10 (HPJ 11-HPJ $10^*$) were chosen as candidates based on the linearity of the standard curve and the sensitivity of the assay. To further characterize these two pairs. MAbs against human AFP were purified from hybridoma cells. conjugated with HRP. and then characterized to optimize the two-site sandwich ELISA The HPJ 10-HPJ $1^*$ pair showed a sensitivity of 1 ng/ml and a better reproducibility than the HPJ 11-HPJ $10^*$ pair when the human sera were incubated at $37^{\circ}C$ for 30 min. The results obtained for 480 randomly selected human sera showed 0~20 ng/ml of AFP values for the normal human sera. To test the utility of our kit, AFP concentrations were determined for 951 human sera (including 85 normal sera, 480 random blood sera, 213 HBsAg-positives. 50 anti-HCV antibody positives. and 47 malignant diseases) and compared with other commercially available AFP detecting kits. These results show that the present two-site sandwich ELISA method is a rapid, sensitive, and reliable procedure for detecting AFP in human serum.

• 감성과학
• /
• 제8권1호
• /
• pp.1-7
• /
• 2005

This paper reports several emotion detecting methods based on various attributes of human voice. These methods have been developed at our Engineering Systems Laboratory. It is noted that, in all of the proposed methods, only prosodic information in voice is used for emotion recognition and semantic information in voice is not used. Different types of neural networks(NNs) are used for detection depending on the type of voice parameters. Earlier approaches separately used linear prediction coefficients(LPCs) and time series data of pitch but they were combined in later studies. The proposed methods are explained first and then evaluation experiments of individual methods and their performances in emotion detection are presented and compared.

• International Journal of Internet, Broadcasting and Communication
• /
• 제15권3호
• /
• pp.94-102
• /
• 2023

Anomaly detection in human movements can improve safety in indoor workplaces. In this paper, we design a framework for detecting anomalous trajectories of humans in indoor spaces based on a variational autoencoder (VAE) with Bi-LSTM layers. First, the VAE is trained to capture the latent representation of normal trajectories. Then the abnormality of a new trajectory is checked using the trained VAE. In this step, the anomaly score of the trajectory is determined using the trajectory reconstruction error through the VAE. If the anomaly score exceeds a threshold, the trajectory is detected as an anomaly. To select the anomaly threshold, a new metric called D-score is proposed, which measures the difference between recall and precision. The anomaly threshold is selected according to the minimum value of the D-score on the validation set. The MIT Badge dataset, which is a real trajectory dataset of workers in indoor space, is used to evaluate the proposed framework. The experiment results show that our framework effectively identifies abnormal trajectories with 81.22% in terms of the F1-score.