• Journal of Korea Multimedia Society
• /
• v.13 no.5
• /
• pp.722-729
• /
• 2010

Prediction problems are widely used in medical domains. For example, computer aided diagnosis or prognosis is a key component in a CDSS (Clinical Decision Support System). SVMs with nonlinear kernels like RBF kernels, have shown superior accuracy in prediction problems. However, they are not preferred by physicians for medical prediction problems because nonlinear SVMs are difficult to visualize, thus it is hard to provide intuitive interpretation of prediction results to physicians. Nomogram was proposed to visualize SVM classification models. However, it cannot visualize nonlinear SVM models. Localized Radial Basis Function (LRBF) was proposed which shows comparable accuracy as the RBF kernel while the LRBF kernel is easier to interpret since it can be linearly decomposed. This paper presents a new tool named VRIFA, which integrates the nomogram and LRBF kernel to provide users with an interactive visualization of nonlinear SVM models, VRIFA visualizes the internal structure of nonlinear SVM models showing the effect of each feature, the magnitude of the effect, and the change at the prediction output. VRIFA also performs nomogram-based feature selection while training a model in order to remove noise or redundant features and improve the prediction accuracy. The area under the ROC curve (AUC) can be used to evaluate the prediction result when the data set is highly imbalanced. The tool can be used by biomedical researchers for computer-aided diagnosis and risk factor analysis for diseases.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.27 no.5
• /
• pp.619-628
• /
• 2009

LiDAR has capability of rapid data acquisition and provides useful information for reconstructing surface of the Earth. However, Extracting information from LiDAR data is not easy task because LiDAR data consist of irregularly distributed point clouds of 3D coordinates and lack of semantic and visual information. This thesis proposed methods for automatic extraction of buildings and 3D detail modeling using airborne LiDAR data. As for preprocessing, noise and unnecessary data were removed by iterative surface fitting and then classification of ground and non-ground data was performed by analyzing histogram. Footprints of the buildings were extracted by tracing points on the building boundaries. The refined footprints were obtained by regularization based on the building hypothesis. The accuracy of building footprints were evaluated by comparing with 1:1,000 digital vector maps. The horizontal RMSE was 0.56m for test areas. Finally, a method of 3D modeling of roof superstructure was developed. Statistical and geometric information of the LiDAR data on building roof were analyzed to segment data and to determine roof shape. The superstructures on the roof were modeled by 3D analytical functions that were derived by least square method. The accuracy of the 3D modeling was estimated using simulation data. The RMSEs were 0.91m, 1.43m, 1.85m and 1.97m for flat, sloped, arch and dome shapes, respectively. The methods developed in study show that the automation of 3D building modeling process was effectively performed.

• Journal of the Institute of Convergence Signal Processing
• /
• v.19 no.1
• /
• pp.7-13
• /
• 2018

In the current study, we proposed an optimized brain-computer interface (BCI) which employed blind source separation (BBS) approach to remove noises. Thus motor imagery (MI) signal and steady state visual evoked potential (SSVEP) signal were easily to be detected due to enhancement in signal-to-noise ratio (SNR). Moreover, a combination between MI and SSVEP which is typically can increase the number of commands being generated in the current BCI. To reduce the computational time as well as to bring the BCI closer to real-world applications, the current system utilizes a single-channel EEG signal. In addition, a convolutional neural network (CNN) was used as the multi-class classification model. We evaluated the performance in term of accuracy between a non-BBS+BCI and BBS+BCI. Results show that the accuracy of the BBS+BCI is achieved $16.15{\pm}5.12%$ higher than that in the non-BBS+BCI by using BBS than non-used on. Overall, the proposed BCI system demonstrate a feasibility to be applied for multi-dimensional control applications with a comparable accuracy.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.45 no.6
• /
• pp.70-77
• /
• 2008

This paper proposes a new text candidate region detection method that uses genetic algorithm based on information of the segmented regions. In image segmentation, a classification of the pixels at each color channel and a reclassification of the region-unit for reducing inhomogeneous clusters are performed. EWFCM(Entropy-based Weighted C-Means) algorithm to classify the pixels at each color channel is an improved FCM algorithm added with spatial information, and therefore it removes the meaningless regions like noise. A region-based reclassification based on a similarity between each segmented region of the most inhomogeneous cluster and the other clusters reduces the inhomogeneous clusters more efficiently than pixel- and cluster-based reclassifications. And detecting text candidate regions is performed by genetic algorithm based on energy and variance of the directional edge components, the number, and a size of the segmented regions. The region information-based detection method can singles out semantic text candidate regions more accurately than pixel-based detection method and the detection results will be more useful in recognizing the text regions hereafter. Experiments showed the results of the segmentation and the detection. And it confirmed that the proposed method was superior to the existing methods.

• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.10
• /
• pp.447-454
• /
• 2015

This paper proposes geographical name denoising by machine learning of event detection based on twitter. Recently, the increasing number of smart phone users are leading the growing user of SNS. Especially, the functions of short message (less than 140 words) and follow service make twitter has the power of conveying and diffusing the information more quickly. These characteristics and mobile optimised feature make twitter has fast information conveying speed, which can play a role of conveying disasters or events. Related research used the individuals of twitter user as the sensor of event detection to detect events that occur in reality. This research employed geographical name as the keyword by using the characteristic that an event occurs in a specific place. However, it ignored the denoising of relationship between geographical name and homograph, it became an important factor to lower the accuracy of event detection. In this paper, we used removing and forecasting, these two method to applied denoising technique. First after processing the filtering step by using noise related database building, we have determined the existence of geographical name by using the Naive Bayesian classification. Finally by using the experimental data, we earned the probability value of machine learning. On the basis of forecast technique which is proposed in this paper, the reliability of the need for denoising technique has turned out to be 89.6%.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.7
• /
• pp.151-157
• /
• 2019

Recently, monitoring respiration of people has been of interest using non-invasive method. Among the vital signals usually used for indicating health status, non-invasive and portable device based monitoring respiratory status is practically useful and enable one to promptly deal with abnormal physical status. This paper proposes the approach to real-time detection of apnea signal based on Short-Time-Fourier-Transform(STFT). Contrary to the analysis of a signal in frequency domain using Fast-Fourier Transform, this paper employs Short-time-Fourier-Transform so that frequency response can be analyzed in short time interval. The respiratory signal is acquired using UWB radar sensor that enables one to obtain respiration signal in contactless way. Detection of respiratory status is carried out by analyzing frequency response, and classification of respiratory status can be provided. In particular, STFT is employed to analyze respiratory signal in real-time, leading to effective analysis of the respiratory status in practice. In the case of existence of noise in the signal, appropriate filtering process is employed as well. The proposed method is straightforward and is workable in practice to analyze the respiratory status of people. To evaluate the proposed method, experimental results are provided.

• Journal of the Korea Institute of Building Construction
• /
• v.21 no.5
• /
• pp.483-493
• /
• 2021

Urban overcrowding has created an explosive supply and demand for high-rise buildings. High-rise buildings are contributing to enhancing the image of the city by serving as focal points, but due to the stack effect, malfunction of elevator doors, difficulties in opening and closing the doors and windows of the outer wall, smoke and odors spreading to the upper floors, noise, energy loss, fire and pollutants have been causing various unexpected problems such as rapid spread of fire. This study classified high-rise buildings according to their vertical zoning, analyzed the causes of and solutions to the stack effect, and derived design and construction methods. Through the initial plan to block the outside air and securing airtightness through precise construction, we sought ways to secure the airtightness inside and outside the building by actively blocking the airflow from the lower floors. In addition, the facility solution can be a measure to reduce the specific phenomena caused by the stack effect, but it should only be applied to the minimum extent because the potential for secondary damage is high. This study emphasized the need for systematic stack effect management by suggesting design and construction measures for each vertical zoning of the causes and countermeasures of the stack effect. It is expected that this study will be helpful not only for design and construction, but also for building maintenance.

• Journal of Convergence for Information Technology
• /
• v.11 no.6
• /
• pp.33-39
• /
• 2021

Speech recognition technology is being combined with deep learning and is developing at a rapid pace. In particular, voice recognition services are connected to various devices such as artificial intelligence speakers, vehicle voice recognition, and smartphones, and voice recognition technology is being used in various places, not in specific areas of the industry. In this situation, research to meet high expectations for the technology is also being actively conducted. Among them, in the field of natural language processing (NLP), there is a need for research in the field of removing ambient noise or unnecessary voice signals that have a great influence on the speech recognition recognition rate. Many domestic and foreign companies are already using the latest AI technology for such research. Among them, research using a convolutional neural network algorithm (CNN) is being actively conducted. The purpose of this study is to determine the non-voice section from the user's speech section through the convolutional neural network. It collects the voice files (wav) of 5 speakers to generate learning data, and utilizes the convolutional neural network to determine the speech section and the non-voice section. A classification model for discriminating speech sections was created. Afterwards, an experiment was conducted to detect the non-speech section through the generated model, and as a result, an accuracy of 94% was obtained.

• The Journal of the Korea Contents Association
• /
• v.21 no.6
• /
• pp.425-434
• /
• 2021

This study is subjected to two linked research. First is the creation process of dance work applying noise and non-musical sounds in our daily life based on understanding the classification, perception, and factor of the sound, through Murray Schafer's concept of 'soundscape'. Second is to find the value of new type of choreography and musical effect of creation process of the dance work . According to methodological research of practice-based research, three stages which is practice, theory and evaluation were accumulated as somatic data, And the analysis was provided a basis by presenting in a figuration, form of the movement and method of specialization with reference to the paper 『space design and forming practice』(2003). As a result, the creation process was able to discover the musical effect of the sound in daily life and new method of choreography, and also find the possibilities that sound could convey the theme of the dance work, the meaning of the movement and the overall atmosphere of the work to audience. In addition, It is expected that will have been made another new creation environment by potential that music has based on concept 'soundscape'.

• Smart Media Journal
• /
• v.9 no.4
• /
• pp.102-108
• /
• 2020

In this paper, we propose a real-time method of detecting hand motions and extracting the signal frame induced by EF(Electric Field) sensors. The signal induced by hand motion includes not only noises caused by various environmental sources as well as sensor's physical placement, but also different initial off-set conditions. Thus, it has been considered as a challenging problem to detect the motion signal and extract the motion frame automatically in real-time. In this study, we remove the PLN(Power Line Noise) using LPF with 10Hz cut-off and successively apply MA(Moving Average) filter to obtain clean and smooth input motion signals. To sense a hand motion, we use two thresholds(positive and negative thresholds) with offset value to detect a starting as well as an ending moment of the motion. Using this approach, we can achieve the correct motion detection rate over 98%. Once the final motion frame is determined, the motion signals are normalized to be used in next process of classification or recognition stage such as LSTN deep neural networks. Our experiment and analysis show that our proposed methods produce better than 98% performance in correct motion detection rate as well as in frame-matching rate.