• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.6
• /
• pp.1275-1282
• /
• 2014

Too much information exists in ubiquitous environment, and therefore it is not easy to obtain the appropriately classified information from the available data set. Decision tree algorithm is useful in the field of data mining or machine learning system, as it is fast and deduces good result on the problem of classification. Sometimes, however, a decision tree may have leaf nodes which consist of only a few or noise data. The decisions made by those weak leaves will not be effective and therefore should be excluded in the decision process. This paper proposes a method using a classifier, UChoo, for solving a classification problem, and suggests an effective method of decision process involving only the important leaves and thereby excluding the noisy leaves. The experiment shows that this method is effective and reduces the erroneous decisions and can be applied when only important decisions should be made.

• Journal of Biosystems Engineering
• /
• v.35 no.6
• /
• pp.443-449
• /
• 2010

A machine vision-based instrument to measure a droplet size spectrum of a spray nozzle was developed and tested to evaluate its accuracy on measuring spray droplet sizes and classifying nozzle sizes. The instrument consisted of a machine vision, light emitting diode (LED) illumination and a desktop computer. The illumination and machine vision were controlled by the computer through a C++ program. The program controlled the machine vision to capture droplet images under controlled illumination, and processed the droplet images to characterize the droplet size distribution of a spray nozzle. An image processing algorithm was developed to improve the accuracy of the system by eliminating random noise and out-of-focus droplets in droplet images while measuring droplet sizes. The instrument measured sizes of the three different balls (254.0, 497.8 and $793.8\;{\mu}m$) and the measurement ranges were $241.2-273.6\;{\mu}m$, $492.9-529.6\;{\mu}m$ and $800.8-824.1\;{\mu}m$ for 254.0-, 497.84- and $793.75-\;{\mu}m$ balls, respectively. Error of the measured droplet mean was less than 3.0 %. Droplet statistics, $D_{V0.1}$, $D_{V0.5}$ and $D_{V0.9}$, of a reference nozzle set were measured, and droplet size spectra of five spray nozzles covering from very fine to extremely coarse were measured to classify spray nozzle sizes. Ninety percent of the classification results of the instrument agreed with manufacturer's classification. A comparison study was carried out between developed and commercial instruments, and measurement results of the developed instrument were within 20 % of commercial instrument results.

• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.3
• /
• pp.289-294
• /
• 2002

A valid speech-sound block can be classified to provide important information for speech recognition. The classification of the speech-sound block comes from the MRA(multi-resolution analysis) property of the DWT(discrete wavelet transform), which is used to reduce the computational time for the pre-processing of speech recognition. The merging algorithm is proposed to extract valid speech-sounds in terms of position and frequency range. It needs some numerical methods for an adaptive DWT implementation and performs unvoiced/voiced classification and denoising. Since the merging algorithm can decide the processing parameters relating to voices only and is independent of system noises, it is useful for extracting valid speech-sounds. The merging algorithm has an adaptive feature for arbitrary system noises and an excellent denoising SNR(signal-to-nolle ratio).

• Korean Journal of Geomatics
• /
• v.2 no.1
• /
• pp.37-46
• /
• 2002

Accurate classification of water area is an preliminary step to accurately analyze the flooded area and damages caused by flood. This step is especially useful for monitoring the region where annually repeating flood is a problem. The accurate estimation of flooded area can ultimately be utilized as a primary source of information for the policy decision. Although SAR (Synthetic Aperture Radar) imagery with its own energy source is sensitive to the water area, its shadow effect similar to the reflectance signature of the water area should be carefully checked before accurate classification. Especially when we want to identify small flood area with mountainous environment, the step for removing shadow effect turns out to be essential in order to accurately classify the water area from the SAR imagery. In this paper, the flood area was classified and monitored using multi-temporal RADARSAT SAR images of Ok-Chun and Bo-Eun located in Chung-Book Province taken in 12th (during the flood) and 19th (after the flood) of August, 1998. We applied several steps of geometric and radiometric calculations to the SAR imagery. First we reduced the speckle noise of two SAR images and then calculated the radar backscattering coefficient $(\sigma^0)$. After that we performed the ortho-rectification via satellite orbit modeling developed in this study using the ephemeris information of the satellite images and ground control points. We also corrected radiometric distortion caused by the terrain relief. Finally, the water area was identified from two images and the flood area is calculated accordingly. The identified flood area is analyzed by overlapping with the existing land use map.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.6
• /
• pp.1353-1364
• /
• 2018

This study proposed a method to detect Cochlodinium polykrikoides red tide pixels in satellite images using a logistic regression model of machine learning technique under Imbalanced data. The spectral profiles extracted from red tide, clear water, and turbid water were used as training dataset. 70% of the entire data set was extracted and used for as model training, and the classification accuracy of the model was evaluated using the remaining 30%. At this time, the white noise was added to the spectral profile of the red tide, which has a relatively small number of data compared to the clear water and the turbid water, and over-sampling was performed to solve the unbalanced data problem. As a result of the accuracy evaluation, the proposed algorithm showed about 94% classification accuracy.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.6
• /
• pp.1333-1342
• /
• 2018

Until now, Electroencephalography(: EEG) has been the most important and convenient method for the diagnosis and treatment of epilepsy. However, it is difficult to identify the wave characteristics of an epileptic EEG signals because it is very weak, non-stationary and has strong background noise. In this paper, we analyse the effect of dimensionality reduction methods on Epileptic EEG feature selection and classification. Three dimensionality reduction methods: Pincipal Component Analysis(: PCA), Kernel Principal Component Analysis(: KPCA) and Linear Discriminant Analysis(: LDA） were investigated. The performance of each method was evaluated by using Support Vector Machine SVM, Logistic Regression(: LR), K-Nearestneighbor(: K-NN), Decision Tree(: DR) and Random Forest(: RF). From the experimental result, PCA recorded 75% of highest accuracy in SVM, LR and K-NN. KPCA recorded 85% of best performance in SVM and K-KNN while LDA achieved 100% accuracy in K-NN. Thus, LDA dimensionality reduction is found to provide the best classification result for epileptic EEG signal.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.1
• /
• pp.76-84
• /
• 2022

Arrhythmia is a condition in which the heart beats abnormally or irregularly, early detection is very important because it can cause dangerous situations such as fainting or sudden cardiac death. However, performance degradation occurs due to personalized differences in ECG signals. In this paper, we propose arrhythmia classification using hybrid combination model of CNN-LSTM. For this purpose, the R wave is detected from noise removed signal and a single bit segment was extracted. It consisted of eight convolutional layers to extract the features of the arrhythmia in detail, used them as the input of the LSTM. The weights were learned through deep learning and the model was evaluated by the verification data. The performance was compared in terms of the accuracy, precision, recall, F1 score through MIT-BIH arrhythmia database. The achieved scores indicate 92.3%, 90.98%, 92.20%, 90.72% in terms of the accuracy, precision, recall, F1 score, respectively.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.11
• /
• pp.1608-1614
• /
• 2022

As competition between universities intensifies due to the recent decrease in the number of students, it is recognized as an essential task of universities to predict students who are underperforming at an early stage and to make various efforts to prevent dropouts. For this, a high-performance model that accurately predicts student performance is essential. This paper proposes a method to improve prediction performance by removing or amplifying abnormal data in a classification prediction model for identifying underperforming students. Existing anomaly data processing methods have mainly focused on deleting or ignoring data, but this paper presents a criterion to distinguish noise from change indicators, and contributes to improving the performance of predictive models by deleting or amplifying data. In an experiment using open learning performance data for verification of the proposed method, we found a number of cases in which the proposed method can improve classification performance compared to the existing method.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.1
• /
• pp.89-98
• /
• 2022

As deep learning technology was applied to various fields, research on adversarial attack techniques, a security problem of deep learning models, was actively studied. adversarial attacks have been mainly studied in the field of images. Recently, they have even developed a complete decision-based attack technique that can attack with just the classification results of the model. However, in the case of the audio field, research is relatively slow. In this paper, we applied several decision-based attack techniques to the audio field and improved state-of-the-art attack techniques. State-of-the-art decision-attack techniques have the disadvantage of requiring many queries for gradient approximation. In this paper, we improve query efficiency by proposing a method of reducing the vector search space required for gradient approximation. Experimental results showed that the attack success rate was increased by 50%, and the difference between original audio and adversarial examples was reduced by 75%, proving that our method could generate adversarial examples with smaller noise.

• Smart Structures and Systems
• /
• v.30 no.6
• /
• pp.613-626
• /
• 2022

Guaranteeing the quality and integrity of structural health monitoring (SHM) data is very important for an effective assessment of structural condition. However, sensory system may malfunction due to sensor fault or harsh operational environment, resulting in multiple types of data anomaly existing in the measured data. Efficiently and automatically identifying anomalies from the vast amounts of measured data is significant for assessing the structural conditions and early warning for structural failure in SHM. The major challenges of current automated data anomaly detection methods are the imbalance of dataset categories. In terms of the feature of actual anomalous data, this paper proposes a data anomaly detection method based on data-level and deep learning technique for SHM of civil engineering structures. The proposed method consists of a data balancing phase to prepare a comprehensive training dataset based on data-level technique, and an anomaly detection phase based on a sophisticatedly designed network. The advanced densely connected convolutional network (DenseNet) and Transformer encoder are embedded in the specific network to facilitate extraction of both detail and global features of response data, and to establish the mapping between the highest level of abstractive features and data anomaly class. Numerical studies on a steel frame model are conducted to evaluate the performance and noise immunity of using the proposed network for data anomaly detection. The applicability of the proposed method for data anomaly classification is validated with the measured data of a practical supertall structure. The proposed method presents a remarkable performance on data anomaly detection, which reaches a 95.7% overall accuracy with practical engineering structural monitoring data, which demonstrates the effectiveness of data balancing and the robust classification capability of the proposed network.