• Smart Structures and Systems
• /
• v.30 no.6
• /
• pp.613-626
• /
• 2022

Guaranteeing the quality and integrity of structural health monitoring (SHM) data is very important for an effective assessment of structural condition. However, sensory system may malfunction due to sensor fault or harsh operational environment, resulting in multiple types of data anomaly existing in the measured data. Efficiently and automatically identifying anomalies from the vast amounts of measured data is significant for assessing the structural conditions and early warning for structural failure in SHM. The major challenges of current automated data anomaly detection methods are the imbalance of dataset categories. In terms of the feature of actual anomalous data, this paper proposes a data anomaly detection method based on data-level and deep learning technique for SHM of civil engineering structures. The proposed method consists of a data balancing phase to prepare a comprehensive training dataset based on data-level technique, and an anomaly detection phase based on a sophisticatedly designed network. The advanced densely connected convolutional network (DenseNet) and Transformer encoder are embedded in the specific network to facilitate extraction of both detail and global features of response data, and to establish the mapping between the highest level of abstractive features and data anomaly class. Numerical studies on a steel frame model are conducted to evaluate the performance and noise immunity of using the proposed network for data anomaly detection. The applicability of the proposed method for data anomaly classification is validated with the measured data of a practical supertall structure. The proposed method presents a remarkable performance on data anomaly detection, which reaches a 95.7% overall accuracy with practical engineering structural monitoring data, which demonstrates the effectiveness of data balancing and the robust classification capability of the proposed network.

• Proceedings of the Optical Society of Korea Conference
• /
• 2009.02a
• /
• pp.427-428
• /
• 2009

Optical imaging offers excellent spatio-temporal sensitivity that is unparalleled by any other perfusion based imaging techniques. We used in vivo optical recording of intrinsic signals (ORIS) to map neurovascular hemodynamics of perfusion, oximetry and membrane potential during epileptic events in rat and mouse neocortex. Studies of hemodynamic changes with ORIS alone were also performed in human. Laboratory studies in rodent epilepsy models have demonstrated a persistent increase in deoxygenated hemoglobin (Hbr) and a decrease in tissue oxygenation during interictal spikes and ictal events. This "epileptic dip", like the "initial dip" recorded during normal sensory processing, implies that the enormous rise in cerebral blood flow (CBF) is inadequate to meet the increased metabolic demands associated with synchronized epileptic activity. These findings are critically important to the interpretation of the perfusion-based imaging studies, such as fMRI. In addition, we visualized the effect of direct cortical electrical stimulation, an alterative epilepsy treatment. The optical data following direct cortical electrical stimulation showed that hemodynamic signals are sensitive to different electrical stimulation parameters. Furthermore, our recent data demonstrated that the application of unilateral electrical stimulation is able to elicit bilateral hemodynamic responses in rat neocortex.

• Journal of the Korean Society for Precision Engineering
• /
• v.20 no.11
• /
• pp.79-90
• /
• 2003

In this paper, we propose a multisensor-based navigation algorithm for a mobile robot, which is intelligently searching the goal location in unknown dynamic environments using multi-ultrasonic sensor. Instead of using “sensor fusion” method which generates the trajectory of a robot based upon the environment model and sensory data, “command fusion” method by fuzzy inference is used to govern the robot motions. The major factors for robot navigation are represented as a cost function. Using the data of the robot states and the environment, the weight value of each factor using fuzzy inference is determined for an optimal trajectory in dynamic environments. For the evaluation of the proposed algorithm, we performed simulations in PC as well as experiments with IRL-2002. The results show that the proposed algorithm is apt to identify obstacles in unknown environments to guide the robot to the goal location safely.

• Journal of the Korean Society of Industry Convergence
• /
• v.21 no.3
• /
• pp.125-131
• /
• 2018

In this paper, we propose a homogeneous multisensor-based navigation algorithm for a mobile robot, which is intelligently searching the goal location in unknown dynamic environments with moving obstacles using multi-ultrasonic sensor. Instead of using "sensor fusion" method which generates the trajectory of a robot based upon the environment model and sensory data, "command fusion" method by fuzzy inference is used to govern the robot motions. The major factors for robot navigation are represented as a cost function. Using the data of the robot states and the environment, the weight value of each factor using fuzzy inference is determined for an optimal trajectory in dynamic environments. For the evaluation of the proposed algorithm, we performed simulations in PC as well as real experiments with mobile robot, AmigoBot. The results show that the proposed algorithm is apt to identify obstacles in unknown environments to guide the robot to the goal location safely.

• Journal of Information Processing Systems
• /
• v.5 no.2
• /
• pp.41-68
• /
• 2009

Face recognition presents a challenging problem in the field of image analysis and computer vision, and as such has received a great deal of attention over the last few years because of its many applications in various domains. Face recognition techniques can be broadly divided into three categories based on the face data acquisition methodology: methods that operate on intensity images; those that deal with video sequences; and those that require other sensory data such as 3D information or infra-red imagery. In this paper, an overview of some of the well-known methods in each of these categories is provided and some of the benefits and drawbacks of the schemes mentioned therein are examined. Furthermore, a discussion outlining the incentive for using face recognition, the applications of this technology, and some of the difficulties plaguing current systems with regard to this task has also been provided. This paper also mentions some of the most recent algorithms developed for this purpose and attempts to give an idea of the state of the art of face recognition technology.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.527-530
• /
• 1996

The Kalman filter has been used as a self-localization method for the mobile robot. To satisfy the assumptions inherent in the Kalman filter, we should calibrate the sensors of the robot before use of them. However, it is generally hard to find exact sensor parameters, and the parameters may change during the robot task as the environment varies. Thus we need to perform on-line sensor calibration, by which we can obtain more credible location of the mobile robot. In this paper, we present an on-line sensor calibration scheme which estimates the unknown sensor bias and the current position of the robot. To this end, first we find out the calibration errors of the sensor from redundant sensory data using the parity vector and recursive minimum variance estimation. Then we calculate the current position of the robot by weighted least square estimation without internal encoder data. The performance of the proposed method is evaluated through computer simulation.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.26 no.7
• /
• pp.30-41
• /
• 1989

In this paper, a robust and simple structured light sensory system has been studied to endow mobile robots with the ability of navigating in real world. A mobile robot with this sensor can be applied in two ways: first, real time navigation in 3-dimensional world, second, modeling and recognition of environment. Range data obtained with this sensor are fairy accurate, and the data aquisition speed is satisfactory. Experiments in diverse situation show effectiveness of the structured light sensor for the mobile robot.

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2006.04a
• /
• pp.40-54
• /
• 2006

It's not easy to detect the high quality and functional compounds for control quality of food materials. The electronic nose was an instrument, which comprised of an array of electronic chemical sensors with partial specificity and an appropriate pattern recognition system, capable of recognizing simple or complex odors. It can conduct fast analysis and provide simple and straightforward results and is best suited for quality control and process monitoring in the field of functional foods. Numbers of applications of an electronic nose in the functional food industry include discrimination of habitats for medicinal food materials, monitoring storage process, lipid oxidation, and quality control of food and/or processing with principal component analysis, neural network analysis and the electronic nose based on GC-SAW sensor. The electronic nose would be possibly useful for a wide variety of quality control in the functional food and plant cultivation when correlating traditional analytical instrumental data with sensory evaluation results or electronic nose data.

• The Research Journal of the Costume Culture
• /
• v.9 no.1
• /
• pp.86-99
• /
• 2001

The purpose of this study was to identify and profile store browsers in terms of their browsing motives, fashion behavioral characteristics, buying behavior and preferred store attributes. The data were collected through questionnaire from 302 female college students by convenient sampling method. Statistical analysis of factor analysis, x²-test, and t-test were performed in analyzing the data. The browsing motives of browsers were to obtains fashion information, sensory stimulation and diversion from routine life. They showed the high level of fashion involvement, shopping confidence, shopping innovativeness, shopping opinion leadership as well s fashion opinion leadership. Browsers tended to be impulse buyers and spent more money on clothing than non-browsers. The attributes that influence their store choice were the variety of products and brands, information availability ,and pleasant store atmosphere.

• International Journal of Precision Engineering and Manufacturing
• /
• v.4 no.4
• /
• pp.57-63
• /
• 2003

Machining information such as form accuracy and surface roughness is an important factor for manufacturing precise parts. To this regard, OMM (On the Machine Measurement) has been researched for last several decades to alternate CMM (Coordinate Measurement Machine) process. In this research, the OMM system with a laser displacement sensor was developed for measuring form accuracy and surface roughness of the machined workpiece on the machine tool. The surface roughness was estimated comparing the sensory signal with the reference data measured from master specimen. Also, form accuracy was determined from the moving averaged raw data. In addition, the geometric error map constructed beforehand using the geometric errors of the machine tool was used to compensate the obtained form accuracy. The overall performance was compared with CMM result, and verified the feasibility of the measurement system.