• Journal of information and communication convergence engineering
• /
• v.22 no.3
• /
• pp.256-266
• /
• 2024

Sasang constitutional medicine (SCM) is one of the best traditional therapeutic approaches used in Korea. SCM prioritizes personalized treatment that considers the unique constitution of an individual and encompasses their physical characteristics, personality traits, and susceptibility to specific diseases. Facial features are essential for diagnosing Sasang constitutional types (SCTs). This study aimed to develop a real-time artificial intelligence-based model for diagnosing SCTs using facial images, building an SCTs prediction model based on a machine learning method. Facial features from all images were extracted to develop this model using feature engineering and machine learning techniques. The fusion of these features was used to train the AI model. We used four machine learning algorithms, namely, random forest (RF), multilayer perceptron (MLP), gradient boosting machine (GBM), and extreme gradient boosting (XGB), to investigate SCTs. The GBM outperformed all the other models. The highest accuracy achieved in the experiment was 81%, indicating the robustness of the proposed model and suitability for real-time applications.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.1-16
• /
• 2022

Despite recent breakthroughs in deep learning and computer vision fields, the pixel-wise identification of tiny objects in high-resolution images with complex disturbances remains challenging. This study proposes a modified U-net for tiny crack segmentation in real-world steel-box-girder bridges. The modified U-net adopts the common U-net framework and a novel Self-Attention-Self-Adaption (SASA) neuron as the fundamental computing element. The Self-Attention module applies softmax and gate operations to obtain the attention vector. It enables the neuron to focus on the most significant receptive fields when processing large-scale feature maps. The Self-Adaption module consists of a multiplayer perceptron subnet and achieves deeper feature extraction inside a single neuron. For data augmentation, a grid-based crack random elastic deformation (CRED) algorithm is designed to enrich the diversities and irregular shapes of distributed cracks. Grid-based uniform control nodes are first set on both input images and binary labels, random offsets are then employed on these control nodes, and bilinear interpolation is performed for the rest pixels. The proposed SASA neuron and CRED algorithm are simultaneously deployed to train the modified U-net. 200 raw images with a high resolution of 4928 × 3264 are collected, 160 for training and the rest 40 for the test. 512 × 512 patches are generated from the original images by a sliding window with an overlap of 256 as inputs. Results show that the average IoU between the recognized and ground-truth cracks reaches 0.409, which is 29.8% higher than the regular U-net. A five-fold cross-validation study is performed to verify that the proposed method is robust to different training and test images. Ablation experiments further demonstrate the effectiveness of the proposed SASA neuron and CRED algorithm. Promotions of the average IoU individually utilizing the SASA and CRED module add up to the final promotion of the full model, indicating that the SASA and CRED modules contribute to the different stages of model and data in the training process.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.45 no.5
• /
• pp.1-8
• /
• 2008

The belief propagation method that has been studied recently yields good performance in disparity extraction. The method in which a target function is modeled as an energy function based on Markov random field(MRF), solves the stereo matching problem by finding the disparity to minimize the energy function. MRF models provide robust and unified framework for vision problem such as stereo and image restoration. the belief propagation method produces quite correct results, but it has difficulty in real time implementation because of higher computational complexity than other stereo methods. To relieve this problem, in this paper, we propose a fast algorithm of the belief propagation method. Energy function consists of a data term and a smoothness tern. The data term usually corresponds to the difference in brightness between correspondences, and smoothness term indicates the continuity of adjacent pixels. Smoothness information is created from messages, which are assigned using four different message arrays for the pixel positions adjacent in four directions. The processing time for four message arrays dominates 80 percent of the whole program execution time. In the proposed method, we propose an algorithm that dramatically reduces the processing time require in message calculation, since the message.; are not produced in four arrays but in a single array. Tn the last step of disparity extraction process, the messages are called in the single integrated array and this algorithm requires 1/4 computational complexity of the conventional method. Our method is evaluated by comparing the disparity error rates of our method and the conventional method. Experimental results show that the proposed method remarkably reduces the execution time while it rarely increases disparity error.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.10
• /
• pp.133-153
• /
• 2023

In this paper, we propose a novel approach to investigating brain-signal measurement technology using Electroencephalography (EEG). Traditionally, researchers have combined EEG signals with bio-signals (BSs) to enhance the classification performance of emotional states. Our objective was to explore the synergistic effects of coupling EEG and BSs, and determine whether the combination of EEG+BS improves the classification accuracy of emotional states compared to using EEG alone or combining EEG with pseudo-random signals (PS) generated arbitrarily by random generators. Employing four feature extraction methods, we examined four combinations: EEG alone, EG+BS, EEG+BS+PS, and EEG+PS, utilizing data from two widely-used open datasets. Emotional states (task versus rest states) were classified using Support Vector Machine (SVM) and Long Short-Term Memory (LSTM) classifiers. Our results revealed that when using the highest accuracy SVM-FFT, the average error rates of EEG+BS were 4.7% and 6.5% higher than those of EEG+PS and EEG alone, respectively. We also conducted a thorough analysis of EEG+BS by combining numerous PSs. The error rate of EEG+BS+PS displayed a V-shaped curve, initially decreasing due to the deep double descent phenomenon, followed by an increase attributed to the curse of dimensionality. Consequently, our findings suggest that the combination of EEG+BS may not always yield promising classification performance.

• Journal of KIISE
• /
• v.44 no.4
• /
• pp.392-398
• /
• 2017

The point-cloud representation of an object is performed by scanning a space through a laser scanner that is extracting a set of points, and the points are then integrated into the same coordinate system through a registration. The set of the completed registration-integrated point clouds is classified into meaningful regions, shapes, and noises through a mathematical analysis. In this paper, the aim is the matching of a curved area like a cylinder shape in 3D point-cloud data. The matching procedure is the attainment of the center and radius data through the extraction of the cylinder-shape candidates from the sphere that is fitted through the RANdom Sample Consensus (RANSAC) in the point cloud, and completion requires the matching of the curved region with the Catmull-Rom spline from the extracted center-point data using the Principal Component Analysis (PCA). Not only is the proposed method expected to derive a fast estimation result via linear and curved cylinder estimations after a center-axis estimation without constraint and segmentation, but it should also increase the work efficiency of reverse engineering.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.43 no.5 s.311
• /
• pp.37-44
• /
• 2006

In this paper, a new version of the ICM method(MICM, modified ICM) in which the contextual information is modelled by Markov random fields (MRF) is introduced. To extract the feature, a new local MRF model with a fitting block neighbourhood is proposed. This model selects contextual information not only from the relative intensity levels but also from the geometrically directional position of neighbouring cliques. Feature extraction depends on each block's contribution to the local variance. They discriminates it into several regions, for example context and background. Boundaries between these regions are also distinctive. The proposed algerian performs segmentation using directional block fitting procedure which confines merging to spatially adjacent elements and generates a partition such that pixels in unified cluster have a homogeneous intensity level. From experiment with ink rubbed copy images(Takbon, 拓本), this method is determined to be quite effective for feature identification. In particular, the new algorithm preserves the details of the images well without over- and under-smoothing problem occurring in general iterated conditional modes (ICM). And also, it may be noted that this method is applicable to the handwriting recognition.

• Wind and Structures
• /
• v.20 no.6
• /
• pp.763-791
• /
• 2015

"Wind and Ports" is a European project that has been carried out since 2009 to handle wind forecast in port areas through an integrated system made up of an extensive in-situ wind monitoring network, the numerical simulation of wind fields, the statistical analysis of wind climate, and algorithms for medium-term (1-3 days) and short term (0.5-2 hours) wind forecasting. The in-situ wind monitoring network, currently made up of 22 ultrasonic anemometers, provides a unique opportunity for detecting high resolution thunderstorm records and studying their dominant characteristics relevant to wind engineering with special concern for wind actions on structures. In such a framework, the wind velocity of thunderstorms is firstly decomposed into the sum of a slowly-varying mean part plus a residual fluctuation dealt with as a non-stationary random process. The fluctuation, in turn, is expressed as the product of its slowly-varying standard deviation by a reduced turbulence component dealt with as a rapidly-varying stationary Gaussian random process with zero mean and unit standard deviation. The extraction of the mean part of the wind velocity is carried out through a moving average filter, and the effect of the moving average period on the statistical properties of the decomposed signals is evaluated. Among other aspects, special attention is given to the thunderstorm duration, the turbulence intensity, the power spectral density and the integral length scale. Some noteworthy wind velocity ratios that play a crucial role in the thunderstorm loading and response of structures are also analyzed.

• The KIPS Transactions:PartB
• /
• v.11B no.3
• /
• pp.289-296
• /
• 2004

General watermarking techniques tend not to consider intrinsic characteristics of image, so that watermarks are embeded to entire images. In this paper, we present a watermarking algorithm based on wavelet domain, and the watermark is embedded into large coefficients in region of interest(ROI) being based on principle of multi-threshold watermark coding(MTWC) for robust watermark insertion. We try to accomplish both image duality and robustness using human visual system(HVS). The watermarks are embedded in middle frequency bands because the distortion degree of watermarked images appears to be less than lower frequency bands, and the embedded watermarks in the middle bands showed high extraction ratios after some distortion. The watermarks are consisted of pseudo random sequences and detected using Cox's similarity mesurement.

• Korean Journal of Plant Resources
• /
• v.17 no.2
• /
• pp.116-121
• /
• 2004

Panax ginseng discarded and lower than 4th grade is caused by red skin disease showing red color skin in ginseng. This kind of red skin ginseng is found a lot in Panax ginseng rather than Panax quinquefolium, and it is considered that red skin disease might be caused by gene. Therefore, this study was carried out to detect genes resistant to red skin disease using RT-PCR. RNA was extracted from three years old ginseng root of both red skin and normal portion in the same root. After RNA extraction, PCR amplification was performed from cDNA using many random primers. As a result, specific band for red skin was found. It is considered that the gene forming band has possibility to be related with red skin disease, and this gene should be decided if it's related with red skin disease. If that gene is related with red skin disease, it will be used for transformation to foster for resistance to red skin disease as well as for selection marker. Bowever, if it's not related with red skin disease, more primers should be used to find gene related with red skin disease.

• Journal of the Optical Society of Korea
• /
• v.18 no.5
• /
• pp.569-573
• /
• 2014

We demonstrate a low-cost, superbly efficient way of etching for the nano-, and micro-sized pyramid patterns on (100)-oriented Si wafer surfaces for use as a patterned master. We show a way of producing functional optical films for the viewing angle direction brightness-enhancement of Lambertian LED (light emitting diode)/OLED (organic light emitting diode) planar lighting applications. An optimally formulated KOH (Potassium hydroxide) wet etching process enabled random-positioned, and random size-distributed (within a certain size range) pyramid patterns to be developed over the entire (100) silicon wafer substrates up to 8" and a simple replication process of master patterns onto the PC (poly-carbonate) and PMMA (poly-methyl methacrylate) films were performed. Haze ratio values were measured for several film samples exhibiting excellent values over 90% suitable for LED/OLED lighting purposes. Brightness was also improved by 13~14% toward the viewing angle direction. Computational simulations using LightTools$^{TM}$ were also carried out and turned out to be in strong agreement with experimental data. Finally, we could check the feasibility of fabricating low-cost, large area, high performance optical films for commercialization.