• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.9
• /
• pp.1224-1231
• /
• 2018

Automatic age estimation has been used in many social network applications, practical commercial applications, and human-computer interaction visual-surveillance biometrics. However, it has rarely been explored. In this paper, we propose an automatic age estimation system, which includes face detection and convolutional deep learning based on an inception module. The latter is a 22-layer-deep network that serves as the particular category of the inception design. To evaluate the proposed approach, we use 4,000 images of eight different age groups from the Adience age dataset. k-fold cross-validation (k = 5) is applied. A comparison of the performance of the proposed work and recent related methods is presented. The results show that the proposed method significantly outperforms existing methods in terms of the exact accuracy and off-by-one accuracy. The off-by-one accuracy is when the result is off by one adjacent age label to the above or below. For the exact accuracy, the age label of "60+" is classified with the highest accuracy of 76%.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.60 no.4
• /
• pp.877-879
• /
• 2011

Bowel sounds (BS) are produced by the movement of the intestinal contents in the lumen of the gastro-intestinal tract during peristalsis and thus, it can be used clinically as useful indicators of bowel motility. We devised an estimation algorithm of bowel motility based on the regression modeling of the jitter and shimmer of BS signals measured by auscultation. Ten healthy males ($23.5\pm2.1$ years) were examined. Consequently, the correlation coefficient, coefficient of determination and standard error between the colon transit times (CTT) measured by a conventional radiograph and the values estimated by our algorithm were 0.98, 0.96 and 2.86, respectively. Also, through k-fold cross validation, the average value of the absolute differences between them was $5.0\pm2.5$ hours. This method could be used as a complementary tool for the non-invasive measurement of bowel motility.

• Proceedings of the Korean Information Science Society Conference
• /
• 2010.06c
• /
• pp.414-418
• /
• 2010

최근 라이프 로그의 수집과 관리에 관련된 연구가 많이 진행 중에 있다. 또 핸드폰 카메라, 디지털 카메라, 캠코더 등의 발전으로 자신의 일상생활을 비디오로 저장하고, 인터넷을 통해 공유하는 사람도 증가하고 있다. 비디오 데이터는 많은 정보를 포함하고 있는 라이프 로그의 한 예로. 동영상의 촬영 및 수집이 활발해짐에 따라 동영상의 메타정보를 생성하고, 이를 이용해 동영상 검색과 관리에 이용하려는 연구들이 진행 중이다. 본 논문에서는 라이프 로그를 수집하고 수집된 동영상과 라이프 로그를 이용하여 의미정보를 추출하는 시스템을 제안한다. 의미정보란 사용자의 행동을 나타내는 정보로써 컴퓨터 사용, 식사, 집안일, 이동, 외출, 독서, 휴식, 일, 기타로 9가지의 의미정보를 추출한다. 제안하는 방법은 사용자로부터 GPS, 가속도센서, 캠코더를 이용해 실제 데이터를 수집하고, 전처리 과정을 통하여 특징을 추출한다. 이때 추출될 특징은 위치정보와 사용자의 상태정보 그리고 영상처리릍 통한 RGB와 HSL 색공간의 요소와 MPEG-7의 EHD(Edge Histogram Descriptor). CLD(Color Layout Descriptor)이다. 추출된 특징으로부터 사람 행동과 같은 불안정한 상황에서 강점을 보이는 확률모델 네트워크인 베이지안 네트워크를 이용하여 의미정보를 추출한다. 제안하는 방법의 유용성을 보이기 위해 실제 데이터를 수집하고 추론하고 10-Fold Cross-validation을 이용하여 데이터를 검증한다.

• Journal of Information Processing Systems
• /
• v.15 no.3
• /
• pp.694-706
• /
• 2019

In this study, we applied the long short-term memory (LSTM) model to classify the cryptocurrency price time series. We collected historic cryptocurrency price time series data and preprocessed them in order to make them clean for use as train and target data. After such preprocessing, the price time series data were systematically encoded into the three-dimensional price tensor representing the past price changes of cryptocurrencies. We also presented our LSTM model structure as well as how to use such price tensor as input data of the LSTM model. In particular, a grid search-based k-fold cross-validation technique was applied to find the most suitable LSTM model parameters. Lastly, through the comparison of the f1-score values, our study showed that the LSTM model outperforms the gradient boosting model, a general machine learning model known to have relatively good prediction performance, for the time series classification of the cryptocurrency price trend. With the LSTM model, we got a performance improvement of about 7% compared to using the GB model.

• Investigative Magnetic Resonance Imaging
• /
• v.25 no.3
• /
• pp.156-163
• /
• 2021

Recently, neuromelanin and nigrosome imaging techniques have been developed to evaluate the substantia nigra in Parkinson's disease. Previous studies have shown potential benefits of quantitative analysis of neuromelanin and nigrosome images in the substantia nigra, although visual assessments have been performed to evaluate structures in most studies. In this study, we investigate the potential of using deep learning based automatic region segmentation techniques for quantitative analysis of the substantia nigra. The deep convolutional neural network was trained to automatically segment substantia nigra regions on 3D nigrosome and neuromelanin sensitive MR images obtained from 30 subjects. With a 5-fold cross-validation, the mean calculated dice similarity coefficient between manual and deep learning was 0.70 ± 0.11. Although calculated dice similarity coefficients were relatively low due to empirically drawn margins, selected slices were overlapped for more than two slices of all subjects. Our results demonstrate that deep convolutional neural network-based method could provide reliable localization of substantia nigra regions on neuromelanin and nigrosome sensitive MR images.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.26 no.3
• /
• pp.183-191
• /
• 2021

A battery management system (BMS) provides some functions for ensuring safety and reliability that includes algorithms estimating battery states. Given the changes caused by various operating conditions, the state-of-health (SOH), which represents a figure of merit of the battery's ability to store and deliver energy, becomes challenging to estimate. Machine learning methods can be applied to perform accurate SOH estimation. In this study, we propose a Long-Term Recurrent Convolutional Network (LRCN) that combines the Convolutional Neural Network (CNN) and Long Short-term Memory (LSTM) to extract aging characteristics and learn temporal mechanisms. The dataset collected by the battery aging experiments of NASA PCoE is used to train models. The input dataset used part of the charging profile. The accuracy of the proposed model is compared with the CNN and LSTM models using the k-fold cross-validation technique. The proposed model achieves a low RMSE of 2.21%, which shows higher accuracy than others in SOH estimation.

• Imaging Science in Dentistry
• /
• v.52 no.2
• /
• pp.187-195
• /
• 2022

Purpose: This study developed a convolutional neural network (CNN) model to diagnose maxillary sinusitis on panoramic radiographs(PRs) and cone-beam computed tomographic (CBCT) images and evaluated its performance. Materials and Methods: A CNN model, which is an artificial intelligence method, was utilized. The model was trained and tested by applying 5-fold cross-validation to a dataset of 148 healthy and 148 inflamed sinus images. The CNN model was implemented using the PyTorch library of the Python programming language. A receiver operating characteristic curve was plotted, and the area under the curve, accuracy, sensitivity, specificity, positive predictive value, and negative predictive values for both imaging techniques were calculated to evaluate the model. Results: The average accuracy, sensitivity, and specificity of the model in diagnosing sinusitis from PRs were 75.7%, 75.7%, and 75.7%, respectively. The accuracy, sensitivity, and specificity of the deep-learning system in diagnosing sinusitis from CBCT images were 99.7%, 100%, and 99.3%, respectively. Conclusion: The diagnostic performance of the CNN for maxillary sinusitis from PRs was moderately high, whereas it was clearly higher with CBCT images. Three-dimensional images are accepted as the "gold standard" for diagnosis; therefore, this was not an unexpected result. Based on these results, deep-learning systems could be used as an effective guide in assisting with diagnoses, especially for less experienced practitioners.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.6
• /
• pp.1239-1248
• /
• 2021

In this paper, we present FDNN algorithm to perform prediction based on academic understanding. In order to apply prediction based on academic understanding rather than data-dependent prediction to deep learning, we constructed algorithm based on mathematical and hydrology. We construct a model that predicts flow rate of a river as an input of precipitation, and measure the model's performance through K-fold cross validation.

• Geomechanics and Engineering
• /
• v.30 no.1
• /
• pp.75-91
• /
• 2022

This paper aims to show how to use several Machine Learning (ML) methods to estimate the TBM penetration rate systematically (TBM-PR). To this end, 1125 datasets including uniaxial compressive strength (UCS), Brazilian tensile strength (BTS), punch slope index (PSI), distance between the planes of weakness (DPW), orientation of discontinuities (alpha angle-α), rock fracture class (RFC), and actual/measured TBM-PRs were established. To evaluate the ML methods' ability to perform, the 5-fold cross-validation was taken into consideration. Eventually, comparing the ML outcomes and the TBM monitoring data indicated that the ML methods have a very good potential ability in the prediction of TBM-PR. However, the long short-term memory model with a correlation coefficient of 0.9932 and a route mean square error of 2.68E-6 outperformed the remaining six ML algorithms. The backward selection method showed that PSI and RFC were more and less significant parameters on the TBM-PR compared to the others.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.1
• /
• pp.38-59
• /
• 2022

Web services instances can be classified into two categories, namely trusted and untrusted from users. A web service with high throughput (TP) and low response time (RT) instance values is a trusted web service. Web services are not trustworthy due to the mismatch in the guaranteed instance values and the actual values achieved by users. To perform web services selection from users' attained TP and RT values, we need to verify the correct prediction of trusted and untrusted instances from invoked web services. This accurate prediction of web services instances is used to perform the selection of web services. We propose to construct fuzzy rules to label web services instances correctly. This paper presents web services selection using a well-known machine learning algorithm, namely REPTree, for the correct prediction of trusted and untrusted instances. Performance comparison of REPTree with five machine learning models is conducted on web services datasets. We have performed experiments on web services datasets using a ten k-fold cross-validation method. To evaluate the performance of the REPTree classifier, we used accuracy metrics (Sensitivity and Specificity). Experimental results showed that web service (WS1) gained top selection score with the (47.0588%) trusted instances, and web service (WS2) was selected the least with (25.00%) trusted instances. Evaluation results of the proposed web services selection approach were found as (asymptotic sig. = 0.019), demonstrating the relationship between final selection and recommended trust score of web services.