• Journal of the Korean Institute of Telematics and Electronics
• /
• v.25 no.8
• /
• pp.998-1009
• /
• 1988

A Hierarchical Circuit Extract Algotithm, which efficiently extract circuits from VLSI mask pattern information, is programmed. Quad-tree is used as a data structure which includes various CIF circuit elements and instances. This system is composed of CIF input routine, Quad-tree making routine, Transistor finding routine and Connection list making routine. This circuit extractor can extract circuit with hierarchical structure of circuit. This system is designed using YACC and LEX. By programming this algorithm with C language and adopting to various circuits, the effectiveness of this algorithm is showed.

• Proceedings of the IEEK Conference
• /
• 2002.07c
• /
• pp.1827-1830
• /
• 2002

A matrix pencil method for source separation 〔3〕was shown to be an unbiased signal extractor in the presence of temporally white noise. Its efficiency and robustness lies in the fact that the method in 〔3〕 employs only time-delayed correlation matrices of the observation data, In this paper we stress out that the matrix pencil method might suffer from a numerical instability problem, be- cause the symmetric-definite pencil was not exploited. Moreover we present a simple method of constructing a symmetric-definite pencil so that the matrix pencil method is numerically stable.

• Journal of Internet Computing and Services
• /
• v.20 no.4
• /
• pp.91-102
• /
• 2019

In autonomous driving systems, the ability to classify pedestrians in images captured by cameras is very important for pedestrian safety. In the past, after extracting features of pedestrians with HOG(Histogram of Oriented Gradients) or SIFT(Scale-Invariant Feature Transform), people classified them using SVM(Support Vector Machine). However, extracting pedestrian characteristics in such a handcrafted manner has many limitations. Therefore, this paper proposes a method to classify pedestrians reliably and effectively using CNN's(Convolutional Neural Network) deep features and transfer learning. We have experimented with both the fixed feature extractor and the fine-tuning methods, which are two representative transfer learning techniques. Particularly, in the fine-tuning method, we have added a new scheme, called M-Fine(Modified Fine-tuning), which divideslayers into transferred parts and non-transferred parts in three different sizes, and adjusts weights only for layers belonging to non-transferred parts. Experiments on INRIA Person data set with five CNN models(VGGNet, DenseNet, Inception V3, Xception, and MobileNet) showed that CNN's deep features perform better than handcrafted features such as HOG and SIFT, and that the accuracy of Xception (threshold = 0.5) isthe highest at 99.61%. MobileNet, which achieved similar performance to Xception and learned 80% fewer parameters, was the best in terms of efficiency. Among the three transfer learning schemes tested above, the performance of the fine-tuning method was the best. The performance of the M-Fine method was comparable to or slightly lower than that of the fine-tuningmethod, but higher than that of the fixed feature extractor method.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.12 no.5
• /
• pp.787-794
• /
• 2001

In this paper, the development of large-signal model extractor for GaAs FET device through the Monolithic Microwave integrated Circuit(MMIC) is presented. The measurement program controlled by personal computer is developed for the processing of an amount of measured data, and the de-embedding algorithm is added to the program for voltage dropping as attached series resistance on measurement system. The small-signal model parameters are typically consisted of 7 elements that are considered as complexity of large-signal model and its the accuracy of the small-signal model is verified through comparing with measured data as varied bias point. The fitting function model, one of the empirical model, is used for quick simulation. In the process of large-signal model parameter extraction, one-dimensional optimization method is proposed and optimized parameters are extracted. This study can reduce the modeling and measuring time and can secure a suitable model for circuit.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.5
• /
• pp.1037-1042
• /
• 2018

The neuromorphic technology has been researched for decades, which learns and processes the information by imitating the human brain. The hardware implementations of neuromorphic systems are configured with highly parallel processing structures and a number of simple computational units. It can achieve high processing speed, low power consumption, and low hardware complexity. Recently, the interests of the neuromorphic technology for low power and small embedded systems have been increasing rapidly. To implement low-complexity hardware, it is necessary to reduce input data dimension without accuracy loss. This paper proposed a low-complexity artificial intelligent engine which consists of parallel neuron engines and a feature extractor. A artificial intelligent engine has a number of neuron engines and its controller to process multimodal sensor data. We verified the performance of the proposed neuron engine including the designed artificial intelligent engines, the feature extractor, and a Micro Controller Unit(MCU).

• Journal of KIISE:Software and Applications
• /
• v.34 no.7
• /
• pp.681-687
• /
• 2007

Face recognition technique is very popular for a personal information security and user identification in recent years. However, the face recognition system is very hard to be implemented due to the difficulty where change in illumination, pose and facial expression. In this paper, we consider that an illumination change causing the variety of face appearance, virtual image data is generated and added to the D-LDA which was selected as the most suitable feature extractor. A less sensitive recognition system in illumination is represented in this paper. This way that consider nature of several illumination directions generate the virtual training image data that considered an illumination effect of the directions and the change of illumination density. As result of experiences, D-LDA has a less sensitive property in an illumination through ORL, Yale University and Pohang University face database.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.63-75
• /
• 2022

The raw data collected by structural health monitoring (SHM) systems may suffer multiple patterns of anomalies, which pose a significant barrier for an automatic and accurate structural condition assessment. Therefore, the detection and classification of these anomalies is an essential pre-processing step for SHM systems. However, the heterogeneous data patterns, scarce anomalous samples and severe class imbalance make data anomaly detection difficult. In this regard, this study proposes a convolutional neural network-based data anomaly detection method. The time and frequency domains data are transferred as images and used as the input of the neural network for training. ResNet18 is adopted as the feature extractor to avoid training with massive labelled data. In addition, the focal loss function is adopted to soften the class imbalance-induced classification bias. The effectiveness of the proposed method is validated using acceleration data collected in a long-span cable-stayed bridge. The proposed approach detects and classifies data anomalies with high accuracy.

• Journal of Soil and Groundwater Environment
• /
• v.7 no.2
• /
• pp.3-11
• /
• 2002

The objective of this study was to survey the oil contamination around the Mountain Baegun at Uiwang city to obtain the preliminary data for bioremediation. For measuring the oil concentrations and physical properties from soil, we analyzed BTEX. TPH and pH, organic content, water content, pormeability coefficient, gravity, porosity and used the purge & trap method for analyzing BTEX. Using the Accelerated Solvent Extractor, we pretreated the samples and then analyzed TPH using GC-FID as soon as possible. From the analysis results, maximum concentration of TPH was 24.773mg/kg and BTEX was 101.7mg/kg. The results of TPH at the Mountain Baegun were higher than the enforcement standard of soil contamination(Korea) and the BTEX concentrations were also higher than the advisory standard of soil contamination(Korea). From these results, the Mountain Baegun may requires to remedy the oil-contaminated soil. In addition, we performed the field bioremediation test for five weeks at the Mountain Baegun using the microbial additives that were developed by our laboratory. From the results of the field test, we could find the about 95% of the oil was removed from the contaminated soil in five weeks. So we consider that it is the one of the useful solutions to remedy the oil-polluted site.

• Journal of Biomedical Engineering Research
• /
• v.11 no.1
• /
• pp.157-162
• /
• 1990

In this study, a peripheral auditory model is used as a frequency feature extractor and a recurrent network which has recurrent links on input nodes is constructed in order to show the reliability of the recurrent network as a recognizer by executing recognition tests for 4 Korean place names and syllables. In the case of using the general learning rule, it is found that the weights are diverged for a long sequence because of the characteristics of the node function in the hidden and output layers. So, a refined weight compensation method is proposed and, using this method, it is possible to improve the system operation and to use long data. The recognition results are considerably good, even if time worping and endpoint detection are omitted and learning patterns and test patterns are made of average length of data. The recurrent network used in this study reflects well time information of temporal speech signal.

• Journal of the Korean Institute of Telematics and Electronics A
• /
• v.29A no.9
• /
• pp.65-75
• /
• 1992

In this paper, we introduce an integrated layout design system, SOLGER. Our system incorporates useful design tools : a powerful layout editor, a coherent access mechanism for large volumes of design data, an incremental design rule checker for hierarchical design environment, node extractor and electrical rule checker, a technology capture which is used for defining technology-specific information, and a procedural design environment for user customization. Also, we present a modified corner-stitching data structure which allows 45$^{\circ}$-angled bilateral edges. Users are provided with a multi-window design environment and a menu-driven interface. SOLGER is being used for VLSI designs practically.