• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.48 no.5
• /
• pp.64-73
• /
• 2011

The problem of Euclidean minimum spanning tree (EMST) is to connect given nodes in a plane with minimum cost. There are many algorithms for the polynomial time problem as EMST. However, for numerous nodes, the algorithms consume an enormous amount of time to find an optimal solution. In this paper, an approximation scheme using a polynomial time approximation scheme (PTAS) algorithm with dividing and parallel processing for the problem is suggested. This scheme enables to construct a large, approximate EMST within a short duration. Although initially devised for the non-polynomial problem, we employ naive PTAS to construct a vast EMST with dynamic programming. In an experiment, the approximate EMST constructed by the proposed scheme with 15,000 input terminal nodes and 16 partition cells shows 89% and 99% saving in execution time for the serial processing and parallel processing methods, respectively. Therefore, our scheme can be applied to obtain an approximate EMST quickly for numerous input terminal nodes.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.3
• /
• pp.143-150
• /
• 2016

In this paper, we propose Korean monophthong recognition method optimizing muscle mixing based on facial surface EMG signals. We observed that EMG signal patterns and muscle activity may vary according to Korean monophthong pronunciation. We use RMS, VAR, MMAV1, MMAV2 which were shown high recognition accuracy in previous study and Cepstral Coefficients as feature extraction algorithm. And we classify Korean monophthong by QDA(Quadratic Discriminant Analysis) and HMM(Hidden Markov Model). Muscle mixing optimized using input data in training phase, optimized result is applied in recognition phase. Then New data are input, finally Korean monophthong are recognized. Experimental results show that the average recognition accuracy is 85.7% in QDA, 75.1% in HMM.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.12
• /
• pp.145-153
• /
• 2020

Recently, keyword extraction from social big data has been widely used for the purpose of extracting opinions or complaints from the user's perspective. Regarding this, our previous work suggested a method to improve accuracy of keyword extraction based on the notion of cohesion scoring, but its accuracy can be degraded when the number of input reviews is relatively small. This paper presents a method to resolve this issue by applying simplified morphological analysis as a postprocessing step to extracted keywords generated from the algorithm discussed in the previous work. The proposed method enables to add analysis rules necessary to process input data incrementally whenever new data arrives, which leads to reduction of a dictionary size and improvement of analysis efficiency. In addition, an interactive rule adder is provided to minimize efforts to add new rules. To verify performance of the proposed method, experiments were conducted based on real social reviews collected from online, where the results showed that error ratio was reduced from 10% to 1% by applying our method and it took 450 milliseconds to process 5,000 reviews, which means that keyword extraction can be performed in a timely manner in the proposed method.

• Journal of the Korean Society of Industry Convergence
• /
• v.26 no.1
• /
• pp.113-119
• /
• 2023

In this paper, we propose an optimization of a pose estimation deep learning model for recognition of abnormal behavior in unmanned stores using radio frequencies. The radio frequency use millimeter wave in the 30 GHz to 300 GHz band. Due to the short wavelength and strong straightness, it is a frequency with less grayness and less interference due to radio absorption on the object. A millimeter wave radar is used to solve the problem of personal information infringement that may occur in conventional CCTV image-based pose estimation. Deep learning-based pose estimation models generally use convolution neural networks. The convolution neural network is a combination of convolution layers and pooling layers of different types, and there are many cases of convolution filter size, number, and convolution operations, and more cases of combining components. Therefore, it is difficult to find the structure and components of the optimal posture estimation model for input data. Compared with conventional millimeter wave-based posture estimation studies, it is possible to explore the structure and components of the optimal posture estimation model for input data using genetic algorithms, and the performance of optimizing the proposed posture estimation model is excellent. Data are collected for actual unmanned stores, and point cloud data and three-dimensional keypoint information of Kinect Azure are collected using millimeter wave radar for collapse and property damage occurring in unmanned stores. As a result of the experiment, it was confirmed that the error was moored compared to the conventional posture estimation model.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.5
• /
• pp.45-54
• /
• 2009

In this paper, we propose an adult images detection algorithm using a mean intensity filter and an improved 2D Hough Transform. This paper is composed of three major steps including a training step, a recognition step, and a verification step. The training step generates a mean nipple variance filter that will be used for detecting nipple candidate regions in the recognition step. To make the mean variance filter, we converts an input color image into a gray scale image and normalize it, and make an average intensity filter for nipple areas. The recognition step first extracts edge images and finds connected components, and decides nipple candidate regions by considering the ratio of width and height of a connected component. It then decides final nipple candidates by calculating the similarity between the learned nipple average intensity filter and the nipple candidate areas. Also, it detects breast lines of an input image through the improved 2D Hough transform. The verification step detects breast areas and identifies adult images by considering the relations between nipple candidate regions and locations of breast lines.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.26 no.4
• /
• pp.27-39
• /
• 2023

Detecting greenhouses from the remote sensing datasets is useful in identifying the illegal agricultural facilities and predicting the agricultural output of the greenhouses. This research proposed a methodology for automatically detecting greenhouses from a given Planetscope satellite imagery acquired in the areas of Gimje City using the deep learning technique through a series of steps. First, multiple training images with a fixed size that contain the greenhouse features were generated from the five training Planetscope satellite imagery. Next, the YOLO(You Only Look Once) model was trained using the generated training images. Finally, the greenhouse features were detected from the input Planetscope satellite image. Statistical results showed that the 76.4% of the greenhouse features were detected from the input Planetscope satellite imagery by using the trained YOLO model. In future research, the high-resolution satellite imagery with a spatial resolution less than 1m should be used to detect more greenhouse features.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.30B no.1
• /
• pp.29-37
• /
• 1993

The purpose of this paper is to estimate the attitude and the position of SPOT satellite which are needed in producing DEM(Digital Elevation Model) using SPOT satellite image pairs. DEM extraction is consists of three parts. First part is the modeling of satellite position and atitude, second part is the matching of two images to find corresponding point of them and third part is to calculate the elevation of each point by using the result of the first and second part. For modeling inaccessible area, extended modeling algorithm which removes the GCP(Ground Control Point) most errorneous from the GCPs extracted from map iteratively is proposed According to the experiments using a collinearity equation, the second order polynomials are shown to the optimal for .omega.(pitch), and Zs parameters while the first order ones for .kappa.(yaw) .PHI.(roll), Xs, and Ys parameters. The input images used in this paper are 6000*6000 level 1A panchromatic digital SPOT images of Chungchong-do, Korea. With 30 GCPs, experiments on SPOT images show that the planimetric and altimetric RMS errors are 7.11m and 7.10m, respectively, for test points.

• Proceedings of the IEEK Conference
• /
• 2000.11a
• /
• pp.377-380
• /
• 2000

In this paper, a new adaptive linearizer architecture with the predistorter is proposed. In the M.Ghaderi's paper, two analog predistorters and an envelope detector are used. Analog circuits for the analog predistorter and the envelope detector can cause imperfection and inaccuracy of the system and make circuits more complex. To solve those problems, most of processes including the predistortion are made by the DSP. The RLS algorithm is applied so that the errors between power amplifier output signals through the postdistorters and predistorted input signals can be converged to the global minimum.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.4
• /
• pp.60-68
• /
• 1998

In this paper, content addressable momory(CAM) using neural network algorithm is proposed to decrease cell loss and process the large amount of data in streaming mode connectionless server at high speed. To overcome problems of area and power dissipation in look-up table using conventional CAM, the proposed neural network CAM is designed to increase linearly address storage bit about increase of address input bit. Its design and imulation is performed by using VHDL and Compass Tool. Also, its layout is performed by using chip compiler, cell-base P&R tool of compass, in 0.8 .mu.m design rule environment.

• Proceedings of the IEEK Conference
• /
• 2005.11a
• /
• pp.659-662
• /
• 2005

A parameterized FFT/IFFT core generator (PFFT_CoreGen) is designed, which can be used as an essential IP (Intellectual Property) in various OFDM modem designs. The PFFT_CoreGen generates Verilog-HDL models of FFT cores in the range of 64 ${\sim}$ 2048-point. To optimize the performance of the generated FFT cores, the PFFT_CoreGen can select the word-length of input data, internal data and twiddle factors in the range of 8-b ${\sim}$ 24-b. Some design techniques for low-power design are considered from algorithm level to circuit level.