• Journal of Multimedia Information System
• /
• v.9 no.1
• /
• pp.21-32
• /
• 2022

Early-stage diagnosis of Alzheimer's Disease (AD) from Cognitively Normal (CN) patients is crucial because treatment at an early stage of AD can prevent further progress in the AD's severity in the future. Recently, computer-aided diagnosis using magnetic resonance image (MRI) has shown better performance in the classification of AD. However, these methods use a traditional machine learning algorithm that requires supervision and uses a combination of many complicated processes. In recent research, the performance of deep neural networks has outperformed the traditional machine learning algorithms. The ability to learn from the data and extract features on its own makes the neural networks less prone to errors. In this paper, a dense neural network is designed for binary classification of Alzheimer's disease. To create a classifier with better results, we studied result of different activation functions in the prediction. We obtained results from 5-folds validations with combinations of different activation functions and compared with each other, and the one with the best validation score is used to classify the test data. In this experiment, features used to train the model are obtained from the ADNI database after processing them using FreeSurfer software. For 5-folds validation, two groups: AD and CN are classified. The proposed DNN obtained better accuracy than the traditional machine learning algorithms and the compared previous studies for AD vs. CN, AD vs. Mild Cognitive Impairment (MCI), and MCI vs. CN classifications, respectively. This neural network is robust and better.

• Journal of Animal Science and Technology
• /
• v.66 no.1
• /
• pp.167-177
• /
• 2024

Iris segmentation is an initial step for identifying the biometrics of animals when establishing a traceability system for livestock. In this study, we propose a deep learning framework for pixel-wise segmentation of bovine iris with a minimized use of annotation labels utilizing the BovineAAEyes80 public dataset. The proposed image segmentation framework encompasses data collection, data preparation, data augmentation selection, training of 15 deep neural network (DNN) models with varying encoder backbones and segmentation decoder DNNs, and evaluation of the models using multiple metrics and graphical segmentation results. This framework aims to provide comprehensive and in-depth information on each model's training and testing outcomes to optimize bovine iris segmentation performance. In the experiment, U-Net with a VGG16 backbone was identified as the optimal combination of encoder and decoder models for the dataset, achieving an accuracy and dice coefficient score of 99.50% and 98.35%, respectively. Notably, the selected model accurately segmented even corrupted images without proper annotation data. This study contributes to the advancement of iris segmentation and the establishment of a reliable DNN training framework.

• Journal of Broadcast Engineering
• /
• v.24 no.3
• /
• pp.472-484
• /
• 2019

This paper proposes a time-domain sound event detection algorithm using DNN (Deep Neural Network). In this system, time domain sound waveform data which is not converted into the frequency domain is used as input to the DNN. The overall structure uses CRNN structure, and GLU, ResNet, and Squeeze-and-excitation blocks are applied. And proposed structure uses structure that considers features extracted from several layers together. In addition, under the assumption that it is practically difficult to obtain training data with strong labels, this study conducted training using a small number of weakly labeled training data and a large number of unlabeled training data. To efficiently use a small number of training data, the training data applied data augmentation methods such as time stretching, pitch change, DRC (dynamic range compression), and block mixing. Unlabeled data was supplemented with insufficient training data by attaching a pseudo-label. In the case of using the neural network and the data augmentation method proposed in this paper, the sound event detection performance is improved by about 6 %(based on the f-score), compared with the case where the neural network of the CRNN structure is used by training in the conventional method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.5
• /
• pp.703-713
• /
• 2022

Accurate and rapid establishment of mobile communication is important in mobile communication system. Currently, the base station parameters to establish a network are determined by cell planning tool. However, it is necessary to perform new cell planning for each new installation of the base station, and there may be a problem that parameters are not suitable for the actual environment are set, such as obstacle information that is not applied in the cell planning tool. In this paper, we proposed methods for path loss prediction using DNN and topographical division using CNN in SON server. After topography classification, a SON server configures the base station parameters according to topography, and update parameters for each topography. The proposed methods can configure the base station parameters automatically that are considered topography information and environmental changes.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.21 no.2
• /
• pp.85-96
• /
• 2019

Solar radiation is an important variable for estimation of energy balance and water cycle in natural and agricultural ecosystems. A deep neural network (DNN) model has been developed in order to estimate the daily global solar radiation. Temperature and precipitation, which would have wider availability from weather stations than other variables such as sunshine duration, were used as inputs to the DNN model. Five-fold cross-validation was applied to train and test the DNN models. Meteorological data at 15 weather stations were collected for a long term period, e.g., > 30 years in Korea. The DNN model obtained from the cross-validation had relatively small value of RMSE ($3.75MJ\;m^{-2}\;d^{-1}$) for estimates of the daily solar radiation at the weather station in Suwon. The DNN model explained about 68% of variation in observed solar radiation at the Suwon weather station. It was found that the measurements of solar radiation in 1985 and 1998 were considerably low for a small period of time compared with sunshine duration. This suggested that assessment of the quality for the observation data for solar radiation would be needed in further studies. When data for those years were excluded from the data analysis, the DNN model had slightly greater degree of agreement statistics. For example, the values of $R^2$ and RMSE were 0.72 and $3.55MJ\;m^{-2}\;d^{-1}$, respectively. Our results indicate that a DNN would be useful for the development a solar radiation estimation model using temperature and precipitation, which are usually available for downscaled scenario data for future climate conditions. Thus, such a DNN model would be useful for the impact assessment of climate change on crop production where solar radiation is used as a required input variable to a crop model.

• Journal of Broadcast Engineering
• /
• v.25 no.4
• /
• pp.609-619
• /
• 2020

In accordance with a demand for high quality video streaming, it needs high data rate in limited bandwidth and more traffic congestion occurs. In particular, when providing real time video service, packet loss rate and bit error probability increase significantly. To solve these problems, a raptor code, which is one of FEC(Forward Error Correction) techniques, is pervasively used in the application layers as a method for improving real-time service quality. In this paper, we propose a method of determining image transmission parameters based on various deep neural networks to increase transmission efficiency at a similar level of image quality by using raptor codes. The proposed neural network uses the packet loss rate, video encoding rate and data rate as inputs, and outputs raptor FEC parameters and packet sizes. The results of the proposed method present that the throughput is 1.2% higher than that of the existing multimedia transmission technique by optimizing the transmission efficiency at a PSNR(Peak Signal-to-Noise Ratio) level similar to that of the existing technique.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.5
• /
• pp.35-41
• /
• 2017

In this paper, we analyze the performance of the recently introduced Hint-knowledge distillation (KD) training approach based on the teacher-student framework for knowledge distillation and knowledge transfer. As a deep neural network (DNN) considered in this paper, the deep residual network (ResNet), which is currently regarded as the latest DNN, is used for the teacher-student framework. Therefore, when implementing the Hint-KD training, we investigate the impact on the weight of KD information based on the soften factor in terms of classification accuracy using the widely used open deep learning frameworks, Caffe. As a results, it can be seen that the recognition accuracy of the student model is improved when the fixed value of the KD information is maintained rather than the gradual decrease of the KD information during training.

• Journal of Information Processing Systems
• /
• v.20 no.3
• /
• pp.360-374
• /
• 2024

A personalized course recommendation algorithm based on deep learning in an online education cloud platform is proposed to address the challenges associated with effective information extraction and insufficient feature extraction. First, the user potential preferences are obtained through the course summary, course review information, user course history, and other data. Second, by embedding, the word vector is turned into a low-dimensional and dense real-valued vector, which is then fed into the compressed interaction network-deep neural network model. Finally, considering that learners and different interactive courses play different roles in the final recommendation and prediction results, an attention mechanism is introduced. The accuracy, recall rate, and F1 value of the proposed method are 0.851, 0.856, and 0.853, respectively, when the length of the recommendation list K is 35. Consequently, the proposed strategy outperforms the comparison model in terms of recommending customized course resources.

• Phonetics and Speech Sciences
• /
• v.7 no.4
• /
• pp.27-33
• /
• 2015

This paper proposes an audio event classification method using Deep Neural Networks (DNN). The proposed method applies Feed Forward Neural Network (FFNN) to generate event probabilities of ten audio events (dog barks, engine idling, and so on) for each frame. For each frame, mel scale filter bank features of its consecutive frames are used as the input vector of the FFNN. These event probabilities are accumulated for the events and the classification result is determined as the event with the highest accumulated probability. For the same dataset, the best accuracy of previous studies was reported as about 70% when the Support Vector Machine (SVM) was applied. The best accuracy of the proposed method achieves as 79.23% for the UrbanSound8K dataset when 80 mel scale filter bank features each from 7 consecutive frames (in total 560) were implemented as the input vector for the FFNN with two hidden layers and 2,000 neurons per hidden layer. In this configuration, the rectified linear unit was suggested as its activation function.

• 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.