• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.44 no.4 s.316
• /
• pp.36-44
• /
• 2007

In this paper, we implement the speech & face recognition system to support various ubiquitous sensor network application services such as switch control, authentication, etc. using wireless audio and image interface. The proposed system is consist of the H/W with audio and image sensor and S/W such as speech recognition algorithm using psychoacoustic model, face recognition algorithm using PCA (Principal Components Analysis) and LDPC (Low Density Parity Check). The proposed speech and face recognition systems are inserted in a HOST PC to use the sensor energy effectively. And improve the accuracy of speech and face recognition, we implement a FEC (Forward Error Correction) system Also, we optimized the simulation coefficient and test environment to effectively remove the wireless channel noises and correcting wireless channel errors. As a result, when the distance that between audio sensor and the source of voice is less then 1.5m FAR and FRR are 0.126% and 7.5% respectively. The face recognition algorithm step is limited 2 times, GAR and FAR are 98.5% and 0.036%.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.181-193
• /
• 2022

Data-driven structural health monitoring (SHM) of civil infrastructure can be used to continuously assess the state of a structure, allowing preemptive safety measures to be carried out. Long-term monitoring of large-scale civil infrastructure often involves data-collection using a network of numerous sensors of various types. Malfunctioning sensors in the network are common, which can disrupt the condition assessment and even lead to false-negative indications of damage. The overwhelming size of the data collected renders manual approaches to ensure data quality intractable. The task of detecting and classifying an anomaly in the raw data is non-trivial. We propose an approach to automate this task, improving upon the previously developed technique of image-based pre-processing on one-dimensional (1D) data by enriching the features of the neural network input data with multiple channels. In particular, feature engineering is employed to convert the measured time histories into a 3-channel image comprised of (i) the time history, (ii) the spectrogram, and (iii) the probability density function representation of the signal. To demonstrate this approach, a CNN model is designed and trained on a dataset consisting of acceleration records of sensors installed on a long-span bridge, with the goal of fault detection and classification. The effect of imbalance in anomaly patterns observed is studied to better account for unseen test cases. The proposed framework achieves high overall accuracy and recall even when tested on an unseen dataset that is much larger than the samples used for training, offering a viable solution for implementation on full-scale structures where limited labeled-training data is available.

• Proceedings of the Korea Committee for Ocean Resources and Engineering Conference
• /
• 2001.10a
• /
• pp.190-195
• /
• 2001

Flow patterns are very complex and interactive between cylinders. The patterns are turbulent and non-linear caused by various factors. In this paper, flow patterns and pressure gradient around vertical cylinders were investigated by experiment. Changing gaps between cylinders the flow patterns are measured at a fixed coming velocity. Flow patterns showed very complex and closely related to the coming velocity and cylinder space. The pressure gradient around the flow field is observed by twelve hole pitot tubes and manometer. The experiment has been conducted in circulating water channel with PIV system. That can visualize flow patterns. The laser beam was used to reflect the image from particles and recorded by CCD camera. The cylinders were spaced from ID to 5D with 0.5m/sec of incoming flow velocity. The experimental results using pitot tube showed in good agreement with results of precious by others study. The results can be applied in the understanding and design of multiple pile array structures.

• Proceedings of the KSRS Conference
• /
• 2005.10a
• /
• pp.637-640
• /
• 2005

In the development of a electronic system for a optical payload comprising mainly EOS(Electro-Optical Sub-system) and PDTS(Payload Data Transmission Sub-system), many aspects should be investigated and discussed for the easy implementation, for th e higher reliability of operation and for the effective ness in cost, size and weight as well as for the secure interface with components of a satellite bus, etc. As important aspects the interfaces between a satellite bus and a payload, and some design features of the CEU(Camera Electronics Unit) inside the payload are described in this paper. Interfaces between a satellite bus and a payload depend considerably on whether t he payload carries the PMU(Payload Management Un it), which functions as main controller of the Payload, or not. With the PMU inside the payload, EOS and PDTS control is performed through the PMU keep ing the least interfaces of control signals and primary power lines, while the EOS and PDTS control is performed directly by the satellite bus components using relatively many control signals when no PMU exists inside the payload. For the CEU design the output channel configurations of panchromatic and multi-spectral bands including the video image data inter face between EOS and PDTS are described conceptually. The timing information control which is also important and necessary to interpret the received image data is described.

• Proceedings of the IEEK Conference
• /
• 2000.06c
• /
• pp.181-184
• /
• 2000

In this paper, we propose new features for describing texture images by using multi-channel Gabor filter and log-polar transform based on human visual system (HVS). Gabor features are extracted by the mean and standard deviation of energy in Gabor response, followed by Fourier series extension. Log-polar features are extracted by log-polar transform and projection. The proposed texture descriptor performs reasonably well with less number of features than other texture descriptors, which has been verified by experiments using some texture images of MPEG-7 data set.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.11a
• /
• pp.575-578
• /
• 2003

본 논문에서는 얼굴영역을 검출하기위해 얼굴 피부색을 보다 효과적으로 모델링하기 위한 방법으로 피부색 특성을 고려하여 자기 성분을 제거한 Red, Blue, Green 채널을 모두 사용하는 Hue, Cb, Cg의 Multi-Channel 피부색 모델을 사용한다. 얼굴영역을 분리한 영상에 Haar 웨이블릿을 이용한 에지영상 추출과 얼굴영역의 특징벡터를 구하기 위하여 26개의 특징벡터를 사용한 효율적인 고차 국소 자동 상관함수를 사용하였다. 계산된 특징벡터는 BP 신경망의 학습을 통하여 얼굴인식을 위한 데이터로 사용된다. 시뮬레이션을 통해 제안된 알고리즘에 의한 인식률향상과 속도 향상을 입증한다.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.227-228
• /
• 2008

A diagnosis based on EM is composed of multi-channel transceiver, antennas in illumination tank, liquid and image reconstruction algorithm for solving inverse scattering problem. The antennas in diagnosis were fabricated and measured in lossy liquid. The 10dB impedance bandwidth of the antenna is 600MHz - 3.5GHz

• Proceedings of the Korean Society of Marine Engineers Conference
• /
• 2006.06a
• /
• pp.169-170
• /
• 2006

The main purpose of this study is to develop a program for sound field visualization system which gets noise signals in microphones array for incoming noise signals and it uses to operate noise signals and to store data in multi-channel FFT and is consisted to visualize noise signals with a image which is got by camera in center of array by using beamforming algorithm of the array signal processing.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.24 no.6
• /
• pp.1040-1048
• /
• 1987

We propose a hardware architecture called Multiring which is applicable for various geometrical operations on rectilinear objects such as design rule checking in VLSI layout and many image processing operations including noise suppression and coutour extraction. It has both a fast execution speed and extremely high flexibility. The whole architecture is mainly divided into four parts` I/O between host and Multiring, ring memory, linear processor array and instruction decoder. Data transmission between host and Multiring is bit serial thereby reducing the bandwidth requirement for teh channel and the number of external pins, while each row data in the bit map stored in ring memory is processed in the corresponding processor in full parallelism. Each processor is simultaneously configured by the instruction decoder/controller to perform one of the 16 basic instructions such as Boolean (AND, OR, NOT, and Copy), geometrical(Expand and Shrink), and I/O operations each ring cycle, which gives Multiring maximal flexibility in terms of design rule change or the instruction set enhancement. Correct functional behavior of Multiring was confirmed by successfully running a software simulator having one-to-one structural correspondence to the Multiring hardware.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.747-749
• /
• 2003

For forest fire monitoring in relatively cool area like Siberia, design of Decision Support System (DSS) is proposed. The DSS is consisted of three different algorithms to detect potential fires from NOAA AVHRR image. The algorithm developed by CCRS (Canada Center for Remote Sensing) uses fixed thresholds for multi-channel information like one by ESA (European Space Agency). The algorithm of IGBP (International Geosphere Biosphere Program) involves contextual information in deriving fire pixels. CCRS and IGBP algorithms are rather liberal compared to more conservative ESA algorithm. Fire pixel information from the three algorithms is presented to the user. The user considers all these information in making decision about the location fire takes place.