• Imaging Science in Dentistry
• /
• v.54 no.1
• /
• pp.81-91
• /
• 2024

Purpose: The objective of this study was to propose a deep-learning model for the detection of the mandibular canal on dental panoramic radiographs. Materials and Methods: A total of 2,100 panoramic radiographs (PANs) were collected from 3 different machines: RAYSCAN Alpha (n=700, PAN A), OP-100 (n=700, PAN B), and CS8100 (n=700, PAN C). Initially, an oral and maxillofacial radiologist coarsely annotated the mandibular canals. For deep learning analysis, convolutional neural networks (CNNs) utilizing U-Net architecture were employed for automated canal segmentation. Seven independent networks were trained using training sets representing all possible combinations of the 3 groups. These networks were then assessed using a hold-out test dataset. Results: Among the 7 networks evaluated, the network trained with all 3 available groups achieved an average precision of 90.6%, a recall of 87.4%, and a Dice similarity coefficient (DSC) of 88.9%. The 3 networks trained using each of the 3 possible 2-group combinations also demonstrated reliable performance for mandibular canal segmentation, as follows: 1) PAN A and B exhibited a mean DSC of 87.9%, 2) PAN A and C displayed a mean DSC of 87.8%, and 3) PAN B and C demonstrated a mean DSC of 88.4%. Conclusion: This multi-device study indicated that the examined CNN-based deep learning approach can achieve excellent canal segmentation performance, with a DSC exceeding 88%. Furthermore, the study highlighted the importance of considering the characteristics of panoramic radiographs when developing a robust deep-learning network, rather than depending solely on the size of the dataset.

• Journal of Intelligence and Information Systems
• /
• v.18 no.1
• /
• pp.39-57
• /
• 2012

Recently, due to the introduction of high-tech equipment in interactive exhibits, many people's attention has been concentrated on Interactive exhibits that can double the exhibition effect through the interaction with the audience. In addition, it is also possible to measure a variety of audience reaction in the interactive exhibition. Among various audience reactions, this research uses the change of the facial features that can be collected in an interactive exhibition space. This research develops an artificial neural network-based prediction model to predict the response of the audience by measuring the change of the facial features when the audience is given stimulation from the non-excited state. To present the emotion state of the audience, this research uses a Valence-Arousal model. So, this research suggests an overall framework composed of the following six steps. The first step is a step of collecting data for modeling. The data was collected from people participated in the 2012 Seoul DMC Culture Open, and the collected data was used for the experiments. The second step extracts 64 facial features from the collected data and compensates the facial feature values. The third step generates independent and dependent variables of an artificial neural network model. The fourth step extracts the independent variable that affects the dependent variable using the statistical technique. The fifth step builds an artificial neural network model and performs a learning process using train set and test set. Finally the last sixth step is to validate the prediction performance of artificial neural network model using the validation data set. The proposed model is compared with statistical predictive model to see whether it had better performance or not. As a result, although the data set in this experiment had much noise, the proposed model showed better results when the model was compared with multiple regression analysis model. If the prediction model of audience reaction was used in the real exhibition, it will be able to provide countermeasures and services appropriate to the audience's reaction viewing the exhibits. Specifically, if the arousal of audience about Exhibits is low, Action to increase arousal of the audience will be taken. For instance, we recommend the audience another preferred contents or using a light or sound to focus on these exhibits. In other words, when planning future exhibitions, planning the exhibition to satisfy various audience preferences would be possible. And it is expected to foster a personalized environment to concentrate on the exhibits. But, the proposed model in this research still shows the low prediction accuracy. The cause is in some parts as follows : First, the data covers diverse visitors of real exhibitions, so it was difficult to control the optimized experimental environment. So, the collected data has much noise, and it would results a lower accuracy. In further research, the data collection will be conducted in a more optimized experimental environment. The further research to increase the accuracy of the predictions of the model will be conducted. Second, using changes of facial expression only is thought to be not enough to extract audience emotions. If facial expression is combined with other responses, such as the sound, audience behavior, it would result a better result.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.26 no.2
• /
• pp.93-98
• /
• 2016

For accurate precipitation forecasts the choice of weather factors and prediction method is very important. Recently, machine learning has been widely used for forecasting precipitation, and artificial neural network, one of machine learning techniques, showed good performance. In this paper, we suggest a new method for forecasting precipitation using DBN, one of deep learning techniques. DBN has an advantage that initial weights are set by unsupervised learning, so this compensates for the defects of artificial neural networks. We used past precipitation, temperature, and the parameters of the sun and moon's motion as features for forecasting precipitation. The dataset consists of observation data which had been measured for 40 years from AWS in Seoul. Experiments were based on 8-fold cross validation. As a result of estimation, we got probabilities of test dataset, so threshold was used for the decision of precipitation. CSI and Bias were used for indicating the precision of precipitation. Our experimental results showed that DBN performed better than MLP.

• Journal of Wetlands Research
• /
• v.21 no.spc
• /
• pp.61-68
• /
• 2019

Recently, climate indices represented by quantifying atmospheric-ocean circulation patterns have been widely used to predict hydrologic variables for considering long-term climate variability. Hydrologic forecasting models based on artificial neural networks have been developed to provide accurate and stable forecasting performance. Forecasts of hydrologic variables considering climate variability can be effectively used for long-term management of water resources and environmental preservation. Therefore, identifying significant indicators for hydrologic variables and applying forecasting models still remains as a challenge. In this study, we selected representative climate indices that have significant relationships with dam inflow time series in the Han-River basin, South Korea for applying the dam inflow forecasting model. For this purpose, the ensemble empirical mode decomposition(EEMD) method was used to identify a significance between dam inflow and climate indices and an artificial neural network(ANN) ensemble model was applied to overcome the limitation of a single ANN model. As a result, the forecasting performances showed that the mean correlation coefficient of the five dams in the training period is 0.88, and the test period is 0.68. It can be expected to come out various applications using the relationship between hydrologic variables and climate variability in South Korea.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.3
• /
• pp.1255-1262
• /
• 2012

Look into the nuclear power plant of Wolsong currently, it is controlled in order to required operating pressure with PI controller. PI controller has a simple structure and satisfy design requirements to gain setting. However, It is difficult to control without changing the gain from produce changes in parameters such as loss of the valves and the pipes. To solve these problems, the dynamic change of the PI controller gain, or to compensate for the PI controller output is desirable to configure the controller. The aim of this research and development in the parameter variations can be controlled to a stable controller design which is reduced an error and a vibration. Proposed PI/NN control techniques is the PI controller and the neural network controller that combines a parallel and the neural network controller part is compensated output of the controller for changes in the parameters were designed to be robust. To directly evaluate the controller performance can be difficult to test in real processes to reflect the characteristics of the process. Therefore, we develope the simulator model using the real process data and simulation results when compared with the simulated process characteristics that showed changes in the parameters. As a result the PI/NN controller error and was confirmed to reduce vibrations.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.8
• /
• pp.31-37
• /
• 2016

Ranging is divided into a synchronous scheme and an asynchronous scheme according to the presence of global synchronization between nodes. In general, the asynchronous ranging is preferred over synchronous ranging because it does not require an expensive high-precision oscillator for the global synchronization. On the other hand, in a conventional asynchronous ranging scheme, the packets, which are generated by all nodes in a positioning system of a large-scale infrastructure and need to be sent for localization by reference nodes, are considerable, which cause an increase in network traffic as the number of nodes increases. The traffic congestion lowers the throughput of the network leading to a considerable loss of energy. To solve this problem, this paper proposes a ranging scheme, in which virtual transmission slots randomly and discretely selected by a plurality of nodes are used to reduce the overheads needed in synchronizing the nodes, and the ranging is performed asynchronously based on the virtual transmission slots, thereby decreasing the network traffic. In addition, a performance test proved that the proposed ranging scheme was stronger than the TWR and SDS-TWR on an error range, even though the intensity of traffic was very low.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.10B
• /
• pp.939-945
• /
• 2008

Tn this paper we described the process and the results of an implementation of Internet terminal using G.729.1 wideband speech codec for next generation network. For this purpose firstly we chose a high performance RISC application processor having DSP features for speech codec processing and enhanced Multimedia Accelerator(eMMA) function for video codec. In the implementation of this terminal, we used G.729.1 codec recently standardized in ITU-T which is a new scalable speech and audio codec that extends 0.729 speech coding standard. To adopt G.729.1 codec to this terminal we transformed most of the fixed point C codes which require more complexity into assembly codes so as to minimize processing time in the processor. As a result of this work we reduced the execution time of the original C codes about 80% and operated in real time on the terminal. For video we used H.263/MPEG-4 codec which is supported by the eMMA with hardware in the processor. In the SIP call processing test connected to real network we obtained under looms end-to-end delay and 3.8 MOS value measured with PESQ instrument. Besides this terminal operated well with commercial terminals.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.40 no.2
• /
• pp.313-324
• /
• 2015

In this paper, in order to support the construction of a smart home environment of low power consumption, we propose a heterogeneous interface determination engine and architecture. Technology of "smart home" is in the spotlight according to the development of IT technology nowadays. Smart homes are configured with multiple sub-networks, and each sub-network is formed by the smart devices using various communication interfaces. Thus, in the smart home environment, interlocking technology between heterogeneous interfaces is essentially required for supporting communication between different networks. Further, each communication interface is a difference in power consumption, and home smart devices are often operated in 24 hours, especially smart phones and other wireless devices are sensitive to power consumption. Therefore, in order to build a energy efficient home network, It is important to select the appropriate interface to handle traffic depending on the situation. In this paper, we propose "The Heterogeneous Interface Decision Engine and Architecture for constructing of Low Power Home Network," and analyze the performance of the proposed method and verify the validity through experiments on the test bed.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.40 no.5
• /
• pp.449-456
• /
• 2022

This study proposes a deep learning algorithm to predict crop yield using GIS (Geographic Information System) to extract soil properties from Soilgrids and soil suitability class maps. The proposed model modified the structure of a published CNN-RNN (Convolutional Neural Network-Recurrent Neural Network) based crop yield prediction model suitable for the domestic crop environment. The existing model has two characteristics. The first is that it replaces the original yield with the average yield of the year, and the second is that it trains the data of the predicted year. The new model uses the original field value to ensure accuracy, and the network structure has been improved so that it can train only with data prior to the year to be predicted. The proposed model predicted the yield per unit area of autumn cabbage for kimchi by region based on weather, soil, soil suitability classes, and yield data from 1980 to 2020. As a result of computing and predicting data for each of the four years from 2018 to 2021, the error amount for the test data set was about 10%, enabling accurate yield prediction, especially in regions with a large proportion of total yield. In addition, both the proposed model and the existing model show that the error gradually decreases as the number of years of training data increases, resulting in improved general-purpose performance as the number of training data increases.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.9
• /
• pp.1-12
• /
• 2022

Traditional methods of measuring corneal ulcers were difficult to present objective basis for diagnosis because of the subjective judgment of the medical staff through photographs taken with special equipment. In this paper, we propose a method to detect the ulcer area on a pixel basis in corneal ulcer images using a semantic segmentation model. In order to solve this problem, we performed the experiment to detect the ulcer area based on the DeepLab model which has the highest performance in semantic segmentation model. For the experiment, the training and test data were selected and the backbone network of DeepLab model which set as Xception and ResNet, respectively were evaluated and compared the performances. We used Dice similarity coefficient and IoU value as an indicator to evaluate the performances. Experimental results show that when 'crop & resized' images are added to the dataset, it segment the ulcer area with an average accuracy about 93% of Dice similarity coefficient on the DeepLab model with ResNet101 as the backbone network. This study shows that the semantic segmentation model used for object detection also has an ability to make significant results when classifying objects with irregular shapes such as corneal ulcers. Ultimately, we will perform the extension of datasets and experiment with adaptive learning methods through future studies so that they can be implemented in real medical diagnosis environment.