• Journal of the Regional Association of Architectural Institute of Korea
• /
• v.20 no.6
• /
• pp.17-23
• /
• 2018

The purpose of this study is to suggest 100MPa class ultra high strength concrete mix design model applying neural network theory, in order to minimize an effort wasted by trials and errors method until now. Mix design model was applied to each of the 70 data using binary binder, ternary binder and quaternary binder. Then being repeatedly applied to back-propagation algorithm in neural network model, optimized connection weight was gained. The completed mix design model was proved, by analyzing and comparing to value predicted from mix design model and value measured from actual compressive strength test. According to the results of this study, more accurate value could be gained through the mix design model, if error rate decreases with the test condition and environment. Also if content of water and binder, slump flow, and air content of concrete apply to mix design model, more accurate and resonable mix design could be gained.

• The Korean Journal of Physiology and Pharmacology
• /
• v.23 no.2
• /
• pp.131-139
• /
• 2019

Manually reviewing electroencephalograms (EEGs) is labor-intensive and demands automated seizure detection systems. To construct an efficient and robust event detector for experimental seizures from continuous EEG monitoring, we combined spectral analysis and deep neural networks. A deep neural network was trained to discriminate periodograms of 5-sec EEG segments from annotated convulsive seizures and the pre- and post-EEG segments. To use the entire EEG for training, a second network was trained with non-seizure EEGs that were misclassified as seizures by the first network. By sequentially applying the dual deep neural networks and simple pre- and post-processing, our autodetector identified all seizure events in 4,272 h of test EEG traces, with only 6 false positive events, corresponding to 100% sensitivity and 98% positive predictive value. Moreover, with pre-processing to reduce the computational burden, scanning and classifying 8,977 h of training and test EEG datasets took only 2.28 h with a personal computer. These results demonstrate that combining a basic feature extractor with dual deep neural networks and rule-based pre- and post-processing can detect convulsive seizures with great accuracy and low computational burden, highlighting the feasibility of our automated seizure detection algorithm.

• Design & Manufacturing
• /
• v.15 no.3
• /
• pp.1-6
• /
• 2021

An artificial neural network model based on a deep learning algorithm is known to be more accurate than humans in image classification, but there is still a limit in the sense that there needs to be a lot of training data that can be called big data. Therefore, various techniques are being studied to build an artificial neural network model with high precision, even with small data. The transfer learning technique is assessed as an excellent alternative. As a result, the purpose of this study is to develop an artificial neural network system that can classify burr images of light guide plate products with 99% accuracy using transfer learning technique. Specifically, for the light guide plate product, 150 images of the normal product and the burr were taken at various angles, heights, positions, etc., respectively. Then, after the preprocessing of images such as thresholding and image augmentation, for a total of 3,300 images were generated. 2,970 images were separated for training, while the remaining 330 images were separated for model accuracy testing. For the transfer learning, a base model was developed using the NASNet-Large model that pre-trained 14 million ImageNet data. According to the final model accuracy test, the 99% accuracy in the image classification for training and test images was confirmed. Consequently, based on the results of this study, it is expected to help develop an integrated AI production management system by training not only the burr but also various defective images.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.49 no.3
• /
• pp.197-204
• /
• 2021

If a failure or an abnormal maneuver occurs during the flight test of a missile, the missile is deliberately self-destructed so as not to continue the flight. At this time, debris are produced and it is important to estimate the impact area in real-time whether it is out of the safety area. In this paper, we propose a method to estimate the debris dispersion area and falling time in real-time using a Fully-Connected Neural Network (FCNN). We applied the Unscented Transform (UT) to generate a large amount of training data. UT parameters were selected by comparing with Monte-Carlo (MC) simulation to secure reliability. Also, we analyzed the performance of the proposed method by comparing the estimation result of MC.

• Journal of Information Processing Systems
• /
• v.18 no.3
• /
• pp.428-442
• /
• 2022

A reliable automatic passenger counting (APC) system is a key point in transportation related to the efficient scheduling and management of transport routes. In this study, we introduce a lightweight head detection network using deep learning applicable to an embedded system. Currently, object detection algorithms using deep learning have been found to be successful. However, these algorithms essentially need a graphics processing unit (GPU) to make them performable in real-time. So, we modify a Tiny-YOLOv3 network using certain techniques to speed up the proposed network and to make it more accurate in a non-GPU environment. Finally, we introduce an APC system, which is performable in real-time on embedded systems, using the proposed head detection algorithm. We implement and test the proposed APC system on a Samsung ARTIK 710 board. The experimental results on three public head datasets reflect the detection accuracy and efficiency of the proposed head detection network against Tiny-YOLOv3. Moreover, to test the proposed APC system, we measured the accuracy and recognition speed by repeating 50 instances of entering and 50 instances of exiting. These experimental results showed 99% accuracy and a 0.041-second recognition speed despite the fact that only the CPU was used.

• International conference on construction engineering and project management
• /
• 2009.05a
• /
• pp.1324-1327
• /
• 2009

Social infrastructure is the basis of public welfare and should be recognized and managed as important assets. Bridge is one of the most important infrastructures to be managed systematically because the impact of the failure is critical. It is essential to monitor the performance of bridges in order to manage them as an asset. But current analytical methods such as predictive modeling and structural analysis are very complicated and difficult to use in practice. To apply these methods, structural and material condition data collection should be performed in each element of bridge. But it is difficult to collect these detailed data in large numbers and various kinds of bridges. Therefore, it is necessary to collect data of major measurement items and predict the life of bridges roughly with advanced information technologies. When certain measurement items reach predefined limits in the monitoring bridges, precise performance measurement will be done by detailed site measurement. This paper describes the selection of major measurement items that can represent the tendency of bridge life and introduces automated bridge data collection test-bed using wireless sensor network technology. The following will be major parts of this paper: 1) Examining the features of conventional bridge management system and data collection method 2) Mileage concept as a bridge life indicator and measuring method of the indicator 3) Test-bed of automated and real-time based bridge life indicator monitoring system using wireless sensor network

• IEIE Transactions on Smart Processing and Computing
• /
• v.3 no.5
• /
• pp.325-330
• /
• 2014

The era of convergence is being applied to all areas of Information and Communication Technology (ICT). The convergence of broadcasting service and communication service almost occurs on smart devices including smartphone. The smart hybrid Digital Multimedia Broadcasting (DMB) is a typical example of the convergence of broadcasting and wireless communication service. The hybrid mobile broadcasting service can support seamless video, 3D, high quality, and additional data services based on network connection between the broadcasting and wireless network. The gateway and terminal (including apps on the smartphone) take the role of the main components on the hybrid service. This paper presents the service concept, main components structure, the implementation of gateway and terminals, and field test to the urban areas for the mobile hybrid system.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.154-157
• /
• 2003

In order for testing faults of combinatorial logic circuit, the authors have developed a new diagnosis method: "Neural Network (NN) fault diagnosis", based on fm error back propagation functions. This method has proved the capability to test gate faults of wider range including so called SSA (single stuck-at) faults, without assuming neither any set of test data nor diagnosis dictionaries. In this paper, it is further shown that what kind of fault models can be detected in the NN fault diagnosis, and the simply modified one can extend to test delay faults, e.g. logic hazard as long as the delays are confined to those due to gates, not to signal lines.

• Computers and Concrete
• /
• v.14 no.5
• /
• pp.563-575
• /
• 2014

Concrete which is a composite material is frequently used in construction works. Properties and behavior of concrete are significant under the effect of different loading cases. Impact loading which is a sudden dynamic one may have destructive effects on structures. Testing apparatuses are designed to investigate the impact effect on test members. Artificial Neural Network (ANN) is a computational model that is inspired by the structure or functional aspects of biological neural networks. It can be defined as an emulation of biological neural system. In this study, impact parameters as acceleration and impact force values of a reinforced concrete slab are obtained by using a testing apparatus and essential test devices. Afterwards, ANN analysis which is used to model different physical dynamic processes depending on several variables is performed in the numerical part of the study. Finally, test and predicted results are compared and it's seen that ANN analysis is an alternative way to predict the results successfully.

• Proceedings of the Korean Society For Composite Materials Conference
• /
• 2003.10a
• /
• pp.111-114
• /
• 2003

Thermosets are highly cross-linked polymers with a three-dimensional molecular structure. The network structure gives rise to mechanical properties, however, one major drawback of thermosets, which also results from their network structure, is their poor resistance to impact and to crack initiation. In this study, to solve this problem, the reactive thermoplastics such as amine terminated polyetherimide (ATPEI), ATPEI-CTBN-ATPEI(ABA) triblock copolymer, CTBN-ATPEI(AB) diblock copolymer, and carboxyl group modified ATPEI was synthesized, after that blended with epoxy resin, and the carbon fiber reinforced composites were fabricated. The impact load, energy, and delamination were investigated by using drop weight impact test and C-scan test. As a results, the ABA/epoxy blend system showed good impact properties.