• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.8
• /
• pp.958-964
• /
• 2016

In this paper, we propose an i-vector based language recognition system to identify the spoken language of the speaker, which uses multiple discriminative deep neural network (DNN) models analogous to the multi-class support vector machine (SVM) classification system. The proposed model was trained and tested using the i-vectors included in the NIST 2015 i-vector Machine Learning Challenge database, and shown to outperform the conventional language recognition methods such as cosine distance, SVM and softmax NN classifier in open-set experiments.

• Journal of Broadcast Engineering
• /
• v.27 no.6
• /
• pp.840-847
• /
• 2022

In a deep neural network (DNN)-based speech enhancement, using global and local input speech information is closely related to model performance. Recently, a nested U-Net structure that utilizes global and local input data information using multi-scale has bee n proposed. This nested U-Net was also applied to speech enhancement and showed outstanding performance. However, a single skip connection used in nested U-Nets must be modified for the nested structure. In this paper, we propose a multi-level skip connection (MLS) to optimize the performance of the nested U-Net-based speech enhancement algorithm. As a result, the proposed MLS showed excellent performance improvement in various objective evaluation metrics compared to the standard skip connection, which means th at the MLS can optimize the performance of the nested U-Net-based speech enhancement algorithm. In addition, the final proposed m odel showed superior performance compared to other DNN-based speech enhancement models.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.727-730
• /
• 2024

최근 딥러닝 기술이 급속히 발전하며 현대 사회의 다양한 응용분야에서 빠르게 적용되고 있다. 특히 영상 기반의 딥러닝 기술은 자연어 처리와 함께 인공지능 기술의 핵심 연구 분야로 많은 연구가 진행되고 있다. 논문에서는 최근 많은 연구가 진행되고 있는 영상의 의미적 분할 (Semantic Segmentation) 성능을 향상하기 위한 연구를 진행한다. 특히 모델에서 고정밀의 의미적 분할을 수행할 수 있도록 추가적인 정보로써 의사 깊이맵 (Pseudo Depth-Map)을 활용하는 방법을 제안하였다. 더불어, 의사 깊이맵을 모델 상에서 효과적으로 학습시키기 위하여 다중 디코더 모델과 학습 효율을 높이는 학습 스케줄링 전략을 제안한다. 의사 깊이맵과 다중 디코더 모델 기반의 제안 모델은 기존 의미적 분할 모델과 비교하여 iIoU 기준 2%의 성능 향상을 보였다.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.2
• /
• pp.59-76
• /
• 2023

Since the COVID-19 era, the rise in apartment prices has been unconventional. In this uncertain real estate market, price prediction research is very important. In this paper, a model is created to predict the actual transaction price of future apartments after building a vast data set of 870,000 from 2015 to 2020 through data collection and crawling on various real estate sites and collecting as many variables as possible. This study first solved the multicollinearity problem by removing and combining variables. After that, a total of five variable selection algorithms were used to extract meaningful independent variables, such as Forward Selection, Backward Elimination, Stepwise Selection, L1 Regulation, and Principal Component Analysis(PCA). In addition, a total of four machine learning and deep learning algorithms were used for deep neural network(DNN), XGBoost, CatBoost, and Linear Regression to learn the model after hyperparameter optimization and compare predictive power between models. In the additional experiment, the experiment was conducted while changing the number of nodes and layers of the DNN to find the most appropriate number of nodes and layers. In conclusion, as a model with the best performance, the actual transaction price of apartments in 2021 was predicted and compared with the actual data in 2021. Through this, I am confident that machine learning and deep learning will help investors make the right decisions when purchasing homes in various economic situations.

• Journal of the Korea Convergence Society
• /
• v.13 no.1
• /
• pp.23-30
• /
• 2022

The failure of the reefer container causes a great loss of cost, but the current reefer container alarm system is inefficient. Existing studies using simulation data of refrigeration systems exist, but studies using actual operation data of refrigeration containers are lacking. Therefore, this study classified the causes of failure using actual refrigerated container operation data. Data imbalance occurred in the actual data, and the data imbalance problem was solved by comparing the logistic regression analysis with ENN-SMOTE and class weight with the 2-stage algorithm developed in this study. The 2-stage algorithm uses XGboost, LGBoost, and DNN to classify faults and normalities in the first step, and to classify the causes of faults in the second step. The model using LGBoost in the 2-stage algorithm was the best with 99.16% accuracy. This study proposes a final model using a two-stage algorithm to solve data imbalance, which is thought to be applicable to other industries.

• Journal of Korea Multimedia Society
• /
• v.23 no.2
• /
• pp.146-154
• /
• 2020

Negative emotion causes stress and lack of attention concentration. The classification of negative emotion is important to recognize risk factors. To classify emotion status, various methods such as questionnaires and interview are used and it could be changed by personal thinking. To solve the problem, we acquire multi modal bio-signals such as electrocardiogram (ECG), skin temperature (ST), galvanic skin response (GSR) and extract features. The neural network (NN), the deep neural network (DNN), and the deep belief network (DBN) is designed using the multi modal bio-signals to analyze emotion status. As a result, the DBN based on features extracted from ECG, ST and GSR shows the highest accuracy (93.8%). It is 5.7% higher than compared to the NN and 1.4% higher than compared to the DNN. It shows 12.2% higher accuracy than using only single bio-signal (GSR). The multi modal bio-signal acquisition and the deep learning classifier play an important role to classify emotion.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.3
• /
• pp.418-424
• /
• 2018

Detection and classification of underwater targets is an important issue for both military and non-military purposes. Recently, many performance improvements are being reported in the field of pattern recognition with the development of deep learning technology. Among the results, DBN showed good performance when used for pre-training of DNN. In this paper, DBN was used for the classification of underwater targets using active sonar, and the results are compared with that of the conventional BPNN. We synthesized active sonar target signals using 3-dimensional highlight model. Then, features were extracted based on FrFT. In the single aspect based experiment, the classification result using DBN was improved about 3.83% compared with the BPNN. In the case of multi-aspect based experiment, a performance of 95% or more is obtained when the number of observation sequence exceeds three.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.70-73
• /
• 2021

최근 영상 및 비디오 분야에 심층 신경망(DNN, Deep Neural Network)을 사용한 연구가 다양하게 진행됨에 따라 High Dynamic Range (HDR) 이미징 기술에서도 기존의 방법들 보다 우수한 성능을 보이는 심층 신경망 모델들이 등장하였다. 하지만, 심층 신경망을 사용한 방법은 큰 연산량과 많은 GPU 메모리를 사용한다는 문제점이 존재하며, 이는 심층 신경망 기반 기술들의 현실 적용 가능성에 제한이 되고 있다. 이에 본 논문에서는 제한된 연산량과 GPU 메모리 조건에서도 사용 가능한 다중 노출 HDR 경량화 심층 신경망을 제안한다. Kalantari Dataset에 대해 기존 HDR 모델들과의 성능 평가를 진행해 본 결과, PSNR-µ와 PSNR-l 수치에서 GPU 메모리 사용량 대비 우수한 성능을 보임을 확인하였다.

• Journal of Wetlands Research
• /
• v.22 no.2
• /
• pp.106-112
• /
• 2020

Wetlands play an important function and role in hydrological, environmental, and ecological, aspects of the watershed. Water level in wetlands is essential for various analysis such as for the determination of wetland function and its effects on the environment. Since several wetlands are ungauged, research on wetland water level prediction are uncommon. Therefore, this study developed a water level prediction model using multiple regression analysis, principal component regression analysis, artificial neural network, and DNN to predict wetland water level. Geumjeong-Mountain Wetland located in Yangsan-city, Gyeongsangnam-do province was selected as the target area, and the water level measurement data from April 2017 to July 2018 was used as the dependent variable. On the other hand, hydrological and meteorological data were used as independent variables in the study. As a result of evaluating the predictive power, the water level prediction model using DNN was selected as the final model as it showed an RMSE value of 6.359 and an NRMSE value of 18.91%. This research study is believed to be useful especially as a basic data for the development of wetland maintenance and management techniques using the water level of the existing unmeasured points.

• Journal of Biomedical Engineering Research
• /
• v.43 no.6
• /
• pp.424-433
• /
• 2022

As the proportion of non-contact telemedicine increases and the number of electrocardiogram (ECG) data measured using portable ECG monitors increases, the demand for automatic algorithms that can precisely analyze vast amounts of ECG is increasing. Since the P, QRS, and T waves of the ECG have different shapes depending on the location of electrodes or individual characteristics and often have similar frequency components or amplitudes, it is difficult to distinguish P, QRS and T waves and measure each parameter. In order to measure the widths, intervals and areas of P, QRS, and T waves, a new algorithm that recognizes the start and end points of each wave and automatically measures the time differences and amplitudes between each point is required. In this study, the start and end points of the P, QRS, and T waves were measured using six Deep Neural Networks (DNN) that recognize the start and end points of each wave. Then, by synthesizing the results of all DNNs, 12 parameters for ECG characteristics for each heartbeat were obtained. In the ECG waveform of 10 subjects provided by Physionet, 12 parameters were measured for each of 660 heartbeats, and the 12 parameters measured for each heartbeat well represented the characteristics of the ECG, so it was possible to distinguish them from other subjects' parameters. When the ECG data of 10 subjects were combined into one file and analyzed with the suggested algorithm, 10 types of ECG waveform were observed, and two types of ECG waveform were simultaneously observed in 5 subjects, however, it was not observed that one person had more than two types.