• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.136-136
• /
• 2022

This study investigates the possibility of coupling empirical mode decomposition (EMD) for runoff prediction from machine learning (ML) models. Here, support vector regression (SVR) and convolutional neural network (CNN) were considered for ML algorithms. Precipitation (P), minimum temperature (Tmin), maximum temperature (Tmax) and their intrinsic mode functions (IMF) values were used for input variables at a monthly scale from Jan. 1973 to Dec. 2020 in the Grand river basin, Canada. The support vector machine-recursive feature elimination (SVM-RFE) technique was applied for finding the best combination of predictors among input variables. The results show that the proposed method outperformed the individual performance of SVR and CNN during the training and testing periods in the study area. According to the correlation coefficient (R), the EMD-SVR model outperformed the EMD-CNN model in both training and testing even though the CNN indicated a better performance than the SVR before using IMF values. The EMD-SVR model showed higher improvement in R value (38.7%) than that from the EMD-CNN model (7.1%). It should be noted that the coupled models of EMD-SVR and EMD-CNN represented much higher accuracy in runoff prediction with respect to the considered evaluation indicators, including root mean square error (RMSE) and R values.

• The Journal of the Acoustical Society of Korea
• /
• v.26 no.6
• /
• pp.243-249
• /
• 2007

We propose an effective voice-based gender identification using group delay(GD) Generally, features for speech recognition are composed of magnitude information rather than phase information. In our approach, we address a difference between male and female for GD which is a derivative of the Fourier transform phase. Also, we propose a novel way to incorporate the features fusion scheme based on a combination of GD and magnitude information such as mel-frequency cepstral coefficients(MFCC), linear predictive coding (LPC) coefficients, reflection coefficients and formant. The experimental results indicate that GD is effective in discriminating gender and the performance is significantly improved when the proposed feature fusion technique is applied.

• Journal of Information Processing Systems
• /
• v.20 no.4
• /
• pp.558-573
• /
• 2024

Considering factors such as illumination, camera quality variations, and background-specific variations, identifying a face using a smartphone-based facial image capture application is challenging. Face Image Quality Assessment refers to the process of taking a face image as input and producing some form of "quality" estimate as an output. Typically, quality assessment techniques use deep learning methods to categorize images. The models used in deep learning are shown as black boxes. This raises the question of the trustworthiness of the models. Several explainability techniques have gained importance in building this trust. Explainability techniques provide visual evidence of the active regions within an image on which the deep learning model makes a prediction. Here, we developed a technique for reliable prediction of facial images before medical analysis and security operations. A combination of gradient-weighted class activation mapping and local interpretable model-agnostic explanations were used to explain the model. This approach has been implemented in the preselection of facial images for skin feature extraction, which is important in critical medical science applications. We demonstrate that the use of combined explanations provides better visual explanations for the model, where both the saliency map and perturbation-based explainability techniques verify predictions.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.1C
• /
• pp.55-64
• /
• 2006

In this paper, a new web-based remote 3D imaging communication system employing an adaptive matching algorithm is suggested. In the proposed method, feature values are extracted from the stereo image pair through estimation of the disparity and similarities between each pixel of the stereo image. And then, the matching window size for disparity estimation is adaptively selected depending on the magnitude of this feature value. Finally, the detected disparity map and the left image is transmitted into the client region through the network channel. And then, in the client region, right image is reconstructed and intermediate views be synthesized by a linear combination of the left and right images using interpolation in real-time. From some experiments on web based-transmission in real-time and synthesis of the intermediate views by using two kinds of stereo images of 'Joo' & 'Hoon' captured by real camera, it is analyzed that PSNRs of the intermediate views reconstructed by using the proposed transmission scheme are highly measured by 30dB for 'Joo', 27dB for 'Hoon' and the delay time required to obtain the intermediate image of 4 view is also kept to be very fast value of 67.2ms on average, respectively.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.4
• /
• pp.161-166
• /
• 2018

In this paper, we propose a method for the random selection of pooling operations for the regularization and reduction of cross validation in convolutional neural networks. The pooling operation in convolutional neural networks is used to reduce the size of the feature map and for its shift invariant properties. In the existing pooling method, one pooling operation is applied in each pooling layer. Because this method fixes the convolution network, the network suffers from overfitting, which means that it excessively fits the models to the training samples. In addition, to find the best combination of pooling operations to maximize the performance, cross validation must be performed. To solve these problems, we introduce the probability concept into the pooling layers. The proposed method does not select one pooling operation in each pooling layer. Instead, we randomly select one pooling operation among multiple pooling operations in each pooling region during training, and for testing purposes, we use probabilistic weighting to produce the expected output. The proposed method can be seen as a technique in which many networks are approximately averaged using a different pooling operation in each pooling region. Therefore, this method avoids the overfitting problem, as well as reducing the amount of cross validation. The experimental results show that the proposed method can achieve better generalization performance and reduce the need for cross validation.

• Journal of the Ergonomics Society of Korea
• /
• v.35 no.5
• /
• pp.371-381
• /
• 2016

Objective:This study aims to evaluate the features of heart rate variability (HRV) and respiratory signals as indices for a driver's drowsiness and waking status in order to develop the classification model for a driver's drowsiness and waking status using those features. Background: Driver's drowsiness is one of the major causal factors for traffic accidents. This study hypothesized that the application of combined bio-signals to monitor the alertness level of drivers would improve the effectiveness of the classification techniques of driver's drowsiness. Method: The features of three heart rate variability (HRV) measurements including low frequency (LF), high frequency (HF), and LF/HF ratio and two respiratory measurements including peak and rate were acquired by the monotonous car driving simulation experiments using the photoplethysmogram (PPG) and respiration sensors. The experiments were repeated a total of 50 times on five healthy male participants in their 20s to 50s. The classification model was developed by selecting the optimal measurements, applying a binary logistic regression method and performing 3-fold cross validation. Results: The power of LF, HF, and LF/HF ratio, and the respiration peak of drowsiness status were reduced by 38%, 22%, 31%, and 7%, compared to those of waking status, while respiration rate was increased by 3%. The classification sensitivity of the model using both HRV and respiratory features (91.4%) was improved, compared to that of the model using only HRV feature (89.8%) and that using only respiratory feature (83.6%). Conclusion: This study suggests that the classification of driver's drowsiness and waking status may be improved by utilizing a combination of HRV and respiratory features. Application: The results of this study can be applied to the development of driver's drowsiness prevention systems.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.10
• /
• pp.207-216
• /
• 2013

The aim of this study is to design a classifier for sleep stage analysis and select important feature set which shows sleep stage well based on physiological signals during sleep. Sleep has a significant effect on the quality of human life. When people undergo lack of sleep or sleep-related disease, they are likely to reduced concentration and cognitive impairment affects, etc. Therefore, there are a lot of research to analyze sleep stage. In this study, after acquisition physiological signals during sleep, we do pre-processing such as filtering for extracting features. The features are used input for the new combination algorithm using genetic algorithm(GA) and neural networks(NN). The algorithm selects features which have high weights to classify sleep stage. As the result of this study, accuracy of the algorithm is up to 90.26% with electroencephalography(EEG) signal and electrocardiography(ECG) signal, and selecting features are alpha and delta frequency band power of EEG signal and standard deviation of all normal RR intervals(SDNN) of ECG signal. We checked the selected features are well shown that they have important information to classify sleep stage as doing repeating the algorithm. This research could use for not only diagnose disease related to sleep but also make a guideline of sleep stage analysis.

• Information Systems Review
• /
• v.17 no.1
• /
• pp.49-64
• /
• 2015

Customer reviews which include subjective opinions for the product or service in online store have been generated rapidly and their influence on customers has become immense due to the widespread usage of SNS. In addition, a number of studies have focused on opinion mining to analyze the positive and negative opinions and get a better solution for customer support and sales. It is very important to select the key terms which reflected the customers' sentiment on the reviews for opinion mining. We proposed a document-level terms-based sentiment classification model by select in the optimal terms with part of speech tag. SVMs (Support vector machines) are utilized to build a predictor for opinion mining and we used the combination of POS tag and four terms extraction methods for the feature selection of SVM. To validate the proposed opinion mining model, we applied it to the customer reviews on Amazon. We eliminated the unmeaning terms known as the stopwords and extracted the useful terms by using part of speech tagging approach after crawling 80,000 reviews. The extracted terms gained from document frequency, TF-IDF, information gain, chi-squared statistic were ranked and 20 ranked terms were used to the feature of SVM model. Our experimental results show that the performance of SVM model with four POS tags is superior to the benchmarked model, which are built by extracting only adjective terms. In addition, the SVM model based on Chi-squared statistic for opinion mining shows the most superior performance among SVM models with 4 different kinds of terms extraction method. Our proposed opinion mining model is expected to improve customer service and gain competitive advantage in online store.

• Journal of KIISE:Software and Applications
• /
• v.26 no.10
• /
• pp.1204-1218
• /
• 1999

문서영상 검색을 위한 디지털도서관의 대부분은 논문제목과/또는 논문요약으로부터 만들어진 색인에 근거한 제한적인 검색기능을 제공하고 있다. 본 논문에서는 영문 문서영상전체에 대한 검색을 위한 단어 영상 형태 특징기반의 단어검색시스템을 제안한다. 본 논문에서는 검색의 효율성과 정확도를 높이기 위해 1) 기존의 단어검색시스템에서 사용된 특징들을 조합하여 사용하며, 2) 특징의 개수 및 위치뿐만 아니라 특징들의 순서를 포함하여 매칭하는 방법을 사용하며, 3) 특징비교에 의해 검색결과를 얻은 후에 여과목적으로 문자인식을 부분적으로 적용하는 2단계의 검색방법을 사용한다. 제안된 시스템의 동작은 다음과 같다. 문서 영상이 주어지면, 문서 영상 구조가 분석되고 단어 영역들의 조합으로 분할된다. 단어 영상의 특징들이 추출되어 저장된다. 사용자의 텍스트 질의가 주어지면 이에 대응되는 단어 영상이 만들어지며 이로부터 영상특징이 추출된다. 이 참조 특징과 저장된 특징들과 비교하여 유사한 단어를 검색하게 된다. 제안된 시스템은 IBM-PC를 이용한 웹 환경에서 구축되었으며, 영문 문서영상을 이용하여 실험이 수행되었다. 실험결과는 본 논문에서 제안하는 방법들의 유효성을 보여주고 있다. Abstract Most existing digital libraries for document image retrieval provide a limited retrieval service due to their indexing from document titles and/or the content of document abstracts. This paper proposes a word spotting system for full English document image retrieval based on word image shape features. In order to improve not only the efficiency but also the precision of a retrieval system, we develop the system by 1) using a combination of the holistic features which have been used in the existing word spotting systems, 2) performing image matching by comparing the order of features in a word in addition to the number of features and their positions, and 3) adopting 2 stage retrieval strategies by obtaining retrieval results by image feature matching and applying OCR(Optical Charater Recognition) partly to the results for filtering purpose. The proposed system operates as follows: given a document image, its structure is analyzed and is segmented into a set of word regions. Then, word shape features are extracted and stored. Given a user's query with text, features are extracted after its corresponding word image is generated. This reference model is compared with the stored features to find out similar words. The proposed system is implemented with IBM-PC in a web environment and its experiments are performed with English document images. Experimental results show the effectiveness of the proposed methods.

• Journal of Biomedical Engineering Research
• /
• v.42 no.3
• /
• pp.71-79
• /
• 2021

Currently, the naked eyes-based diagnosis of bone metastases on CT images relies on qualitative assessment. For this reason, there is a great need for a state-of-the-art approach that can assess and follow-up the bone metastases with quantitative biomarker. Radiomics can be used as a biomarker for objective lesion assessment by extracting quantitative numerical values from digital medical images. In this study, therefore, we evaluated the clinical applicability of non-invasive and objective bone metastases computer-aided diagnosis using radiomics-based biomarkers in CT. We employed a total of 21 approaches consist of three-classifiers and seven-feature selection methods to predict bone metastases and select biomarkers. We extracted three-dimensional features from the CT that three groups consisted of osteoblastic, osteolytic, and normal-healthy vertebral bodies. For evaluation, we compared the prediction results of the classifiers with the medical staff's diagnosis results. As a result of the three-class-classification performance evaluation, we demonstrated that the combination of the random forest classifier and the sequential backward selection feature selection approach reached AUC of 0.74 on average. Moreover, we confirmed that 90-percentile, kurtosis, and energy were the features that contributed high in the classification of bone metastases in this approach. We expect that selected quantitative features will be helpful as biomarkers in improving the patient's survival and quality of life.