• Journal of Intelligence and Information Systems
• /
• v.23 no.1
• /
• pp.47-67
• /
• 2017

Steel plate faults is one of important factors to affect the quality and price of the steel plates. So far many steelmakers generally have used visual inspection method that could be based on an inspector's intuition or experience. Specifically, the inspector checks the steel plate faults by looking the surface of the steel plates. However, the accuracy of this method is critically low that it can cause errors above 30% in judgment. Therefore, accurate steel plate faults diagnosis system has been continuously required in the industry. In order to meet the needs, this study proposed a new steel plate faults diagnosis system using Simultaneous MTS (S-MTS), which is an advanced Mahalanobis Taguchi System (MTS) algorithm, to classify various surface defects of the steel plates. MTS has generally been used to solve binary classification problems in various fields, but MTS was not used for multiclass classification due to its low accuracy. The reason is that only one mahalanobis space is established in the MTS. In contrast, S-MTS is suitable for multi-class classification. That is, S-MTS establishes individual mahalanobis space for each class. 'Simultaneous' implies comparing mahalanobis distances at the same time. The proposed steel plate faults diagnosis system was developed in four main stages. In the first stage, after various reference groups and related variables are defined, data of the steel plate faults is collected and used to establish the individual mahalanobis space per the reference groups and construct the full measurement scale. In the second stage, the mahalanobis distances of test groups is calculated based on the established mahalanobis spaces of the reference groups. Then, appropriateness of the spaces is verified by examining the separability of the mahalanobis diatances. In the third stage, orthogonal arrays and Signal-to-Noise (SN) ratio of dynamic type are applied for variable optimization. Also, Overall SN ratio gain is derived from the SN ratio and SN ratio gain. If the derived overall SN ratio gain is negative, it means that the variable should be removed. However, the variable with the positive gain may be considered as worth keeping. Finally, in the fourth stage, the measurement scale that is composed of selected useful variables is reconstructed. Next, an experimental test should be implemented to verify the ability of multi-class classification and thus the accuracy of the classification is acquired. If the accuracy is acceptable, this diagnosis system can be used for future applications. Also, this study compared the accuracy of the proposed steel plate faults diagnosis system with that of other popular classification algorithms including Decision Tree, Multi Perception Neural Network (MLPNN), Logistic Regression (LR), Support Vector Machine (SVM), Tree Bagger Random Forest, Grid Search (GS), Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). The steel plates faults dataset used in the study is taken from the University of California at Irvine (UCI) machine learning repository. As a result, the proposed steel plate faults diagnosis system based on S-MTS shows 90.79% of classification accuracy. The accuracy of the proposed diagnosis system is 6-27% higher than MLPNN, LR, GS, GA and PSO. Based on the fact that the accuracy of commercial systems is only about 75-80%, it means that the proposed system has enough classification performance to be applied in the industry. In addition, the proposed system can reduce the number of measurement sensors that are installed in the fields because of variable optimization process. These results show that the proposed system not only can have a good ability on the steel plate faults diagnosis but also reduce operation and maintenance cost. For our future work, it will be applied in the fields to validate actual effectiveness of the proposed system and plan to improve the accuracy based on the results.

• Progress in Medical Physics
• /
• v.23 no.4
• /
• pp.317-325
• /
• 2012

The Varian PORTALVISION (Varian Medical Systems, US) shows significant overresponses as the off-center distance increases compared to the predicted dose. In order to correct the dose discrepancy, the off-axis correction is applied to VARIAN iX linear accelerators. The portal dose for $38{\times}28cm^2$ open field is acquired for 6 MV, 15 MV photon beams and also are predicted by PDIP algorithm under the same condition of the portal dose acquisition. The off-axis correction is applied by modifying the $40{\times}40cm^2$ diagonal beam profile data which is used for the beam profile calibration. The ratios between predicted dose and measured dose is modeled as a function of off-axis distance with the $4^{th}$ polynomial and is applied to the $40{\times}40cm^2$ diagonal beam profile data as the weight to correct measured dose by EPID detector. The discrepancy between measured dose and predicted dose is reduced from $4.17{\pm}2.76$ CU to $0.18{\pm}0.8$ CU for 6 MV photon beam and from $3.23{\pm}2.59$ CU to $0.04{\pm}0.85$ CU for 15 MV photon beam. The passing rate of gamma analysis for the pyramid fluence patten with the 4%, 4 mm criteria is improved from 98.7% to 99.1% for 6 MV photon beam, from 99.8% to 99.9% for 15 MV photon beam. IMRT QA is also performed for randomly selected Head and Neck and Prostate IMRT plans after applying the off-axis correction. The gamma passing rare is improved by 3% on average, for Head and Neck cases: $94.7{\pm}3.2%$ to $98.2{\pm}1.4%$, for Prostate cases: $95.5{\pm}2.6%$, $98.4{\pm}1.8%$. The gamma analysis criteria is 3%, 3 mm with 10% threshold. It is considered that the off-axis correction might be an effective and easily adaptable means for correcting the discrepancy between measured dose and predicted dose for IMRT QA using EPID in clinic.

• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.21-35
• /
• 2018

Since the introduction of MP3 players, CD recordings have gradually been vanishing, and the music consuming environment of music users is shifting to mobile devices. The introduction of smart devices has increased the utilization of music through music playback, mass storage, and search functions that are integrated into smartphones and tablets. At the time of initial MP3 player supply, the bitrate of the compressed music contents generally was 128 Kbps. However, as increasing of the demand for high quality music, sound quality of 384 Kbps appeared. Recently, music content of FLAC (Free License Audio Codec) format using lossless compression method is becoming popular. The download service of many music sites in Korea has classified by unlimited download with technical protection and limited download without technical protection. Digital Rights Management (DRM) technology is used as a technical protection measure for unlimited download, but it can only be used with authenticated devices that have DRM installed. Even if music purchased by the user, it cannot be used by other devices. On the contrary, in the case of music that is limited in quantity but not technically protected, there is no way to enforce anyone who distributes it, and in the case of high quality music such as FLAC, the loss is greater. In this paper, the author proposes an audio watermarking technology for copyright protection of high quality stereo music. Two kinds of information, "Copyright" and "Copy_free", are generated by using the turbo code. The two watermarks are composed of 9 bytes (72 bits). If turbo code is applied for error correction, the amount of information to be inserted as 222 bits increases. The 222-bit watermark was expanded to 1024 bits to be robust against additional errors and finally used as a watermark to insert into stereo music. Turbo code is a way to recover raw data if the damaged amount is less than 15% even if part of the code is damaged due to attack of watermarked content. It can be extended to 1024 bits or it can find 222 bits from some damaged contents by increasing the probability, the watermark itself has made it more resistant to attack. The proposed algorithm uses quantization in DCT so that watermark can be detected efficiently and SNR can be improved when stereo music is converted into mono. As a result, on average SNR exceeded 40dB, resulting in sound quality improvements of over 10dB over traditional quantization methods. This is a very significant result because it means relatively 10 times improvement in sound quality. In addition, the sample length required for extracting the watermark can be extracted sufficiently if the length is shorter than 1 second, and the watermark can be completely extracted from music samples of less than one second in all of the MP3 compression having a bit rate of 128 Kbps. The conventional quantization method can extract the watermark with a length of only 1/10 compared to the case where the sampling of the 10-second length largely fails to extract the watermark. In this study, since the length of the watermark embedded into music is 72 bits, it provides sufficient capacity to embed necessary information for music. It is enough bits to identify the music distributed all over the world. 272 can identify $4*10^{21}$, so it can be used as an identifier and it can be used for copyright protection of high quality music service. The proposed algorithm can be used not only for high quality audio but also for development of watermarking algorithm in multimedia such as UHD (Ultra High Definition) TV and high-resolution image. In addition, with the development of digital devices, users are demanding high quality music in the music industry, and artificial intelligence assistant is coming along with high quality music and streaming service. The results of this study can be used to protect the rights of copyright holders in these industries.

• Journal of Intelligence and Information Systems
• /
• v.22 no.2
• /
• pp.127-142
• /
• 2016

Deep learning model is a kind of neural networks that allows multiple hidden layers. There are various deep learning architectures such as convolutional neural networks, deep belief networks and recurrent neural networks. Those have been applied to fields like computer vision, automatic speech recognition, natural language processing, audio recognition and bioinformatics where they have been shown to produce state-of-the-art results on various tasks. Among those architectures, convolutional neural networks and recurrent neural networks are classified as the supervised learning model. And in recent years, those supervised learning models have gained more popularity than unsupervised learning models such as deep belief networks, because supervised learning models have shown fashionable applications in such fields mentioned above. Deep learning models can be trained with backpropagation algorithm. Backpropagation is an abbreviation for "backward propagation of errors" and a common method of training artificial neural networks used in conjunction with an optimization method such as gradient descent. The method calculates the gradient of an error function with respect to all the weights in the network. The gradient is fed to the optimization method which in turn uses it to update the weights, in an attempt to minimize the error function. Convolutional neural networks use a special architecture which is particularly well-adapted to classify images. Using this architecture makes convolutional networks fast to train. This, in turn, helps us train deep, muti-layer networks, which are very good at classifying images. These days, deep convolutional networks are used in most neural networks for image recognition. Convolutional neural networks use three basic ideas: local receptive fields, shared weights, and pooling. By local receptive fields, we mean that each neuron in the first(or any) hidden layer will be connected to a small region of the input(or previous layer's) neurons. Shared weights mean that we're going to use the same weights and bias for each of the local receptive field. This means that all the neurons in the hidden layer detect exactly the same feature, just at different locations in the input image. In addition to the convolutional layers just described, convolutional neural networks also contain pooling layers. Pooling layers are usually used immediately after convolutional layers. What the pooling layers do is to simplify the information in the output from the convolutional layer. Recent convolutional network architectures have 10 to 20 hidden layers and billions of connections between units. Training deep learning networks has taken weeks several years ago, but thanks to progress in GPU and algorithm enhancement, training time has reduced to several hours. Neural networks with time-varying behavior are known as recurrent neural networks or RNNs. A recurrent neural network is a class of artificial neural network where connections between units form a directed cycle. This creates an internal state of the network which allows it to exhibit dynamic temporal behavior. Unlike feedforward neural networks, RNNs can use their internal memory to process arbitrary sequences of inputs. Early RNN models turned out to be very difficult to train, harder even than deep feedforward networks. The reason is the unstable gradient problem such as vanishing gradient and exploding gradient. The gradient can get smaller and smaller as it is propagated back through layers. This makes learning in early layers extremely slow. The problem actually gets worse in RNNs, since gradients aren't just propagated backward through layers, they're propagated backward through time. If the network runs for a long time, that can make the gradient extremely unstable and hard to learn from. It has been possible to incorporate an idea known as long short-term memory units (LSTMs) into RNNs. LSTMs make it much easier to get good results when training RNNs, and many recent papers make use of LSTMs or related ideas.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.05a
• /
• pp.678-681
• /
• 2003

In this paper, we aims to develop a microcontroll er-based portable pulse oximeter using Compact Flash Interface. First, portable pulse oxineter system is designed to record 2 channel of biosignals simultaneously, including 1 channel of SpO$_2$ and 1 channel of pulse rate. It is very small and portable. Besides, the system makes it possible to measure a patients condition without an additional medical equipment. We tried to solve the problems generated by a patient's motion. That is, we added an analog circuit to a traditional pulse oximeter in order to eliminate the change of the base line. And we used 2D sector algorithm. As present, SpO$_2$ modules are completed. But there are still many further development needed in order to enhance the function. Especially, compact flash interface remains the most to complete. Second, ECG monitoring system uses almost same as present 3-lead ECG system. But we focus on the analog part, especially in filter. The proposed filter is composed of two parts. One is a filter to remove the power-line interface. The other is a filter to remove the baseline drift. A filter to remove the power-line and the baseline drift is necessarily used in the ECG system. The implemented filter have three features; minimizing the distortion in DC component, removing the harmonic component of power-line frequency. Using compact flash interface, we can easily transfer a patient's personal information and the measured signal data to a network based server environment. That means, it is possible to implement a patient's monitoring system with low cost.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.43 no.3 s.345
• /
• pp.84-94
• /
• 2006

In this paper, we propose the algorithm to reduce guard time of UWB MAC time slot for throughput gain. In the proposed draft by multiband ofdm alliance (MBOA), Guard time of each medium access slot (MAS) is composed of shortest inter-frame space (SIFS) and MaxDrift which is the time caused by maximum frequency offset among devices. In this paper, to reduceguard time means that we nearly eliminate MaxDrift term from guard time. Each device of a piconet computes relative frequency offset from the device initiating piconet using periodically consecutive transferred beacon frames. Each device add or subtract the calculated relative frequency offset to the estimated each MAS starting point in order to synchronize with calculated MAS starting point of the device initiating piconet. According to verification of simulations, if the frequency offset estimator is implemented with 8 decimal bit, the ratio of the wasted time to Superframe is always less than 0.0001.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.43 no.1 s.343
• /
• pp.57-70
• /
• 2006

This paper descWireless Sensor Network consisting of a number of small sensors with transceiver and data processor is an effective means for gathering data in a variety of environments. The data collected by each sensor is transmitted to a processing center that use all reported data to estimate characteristics of the environment or detect an event. This process must be designed to conserve the limited energy resources of the sensor since neighboring sensors generally have the data of similar information. Therefore, clustering scheme which sends aggregated information to the processing center may save energy. Existing multi-hop cluster energy consumption modeling scheme can not estimate exact energy consumption of an individual sensor. In this paper, we propose a new cluster energy consumption model which modified existing problem. We can estimate more accurate total energy consumption according to the number of clusterheads by using Voronoi tessellation. Thus, we can realize an energy efficient cluster formation. Our modeling has an accuracy over $90\%$ when compared with simulation and has considerably superior than existing modeling scheme about $60\%.$ We also confirmed that energy consumption of the proposed modeling scheme is more accurate when the sensor density is increased.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.46 no.3
• /
• pp.52-58
• /
• 2009

If a moving target has a linear characteristics, the Kalman filter can estimate relatively accurate the location of a target, but this performance depends on how the dynamic status characteristics of the target is accurately modeled. In many practical problems of tracking a maneuvering target, a simple kinematic model can fairly accurately describe the target dynamics for a wide class of maneuvers. However, since the target can exhibit a wide range of dynamic characteristics, no fixed SKF(Simple Kalman filter) can be matched to estimate, to the required accuracy, the states of the target for every specific maneuver. In this paper, a new AKF(Active Kalman filter) is proposed to solve this problem The process noise covariance level of the Kalman filter is adjusted at each time step according to the study result which uses the neural network algorithm. It is demonstrated by means of a computer simulation that the tracking capability of the proposed AKF(Active Kalman filter) is better than that of the SKF(Simple Kalman Filter).

• Journal of Navigation and Port Research
• /
• v.39 no.4
• /
• pp.299-305
• /
• 2015

Modbus is a serial communications protocol, it has since become a practically standard communication protocol, and it is now a commonly available means of connecting industrial electronic devices. Therefore, it can be connected with all devices using Modbus protocol to the measurement and remote control on the ships, buildings, trains, airplanes and etc.. In this paper, we add the Modbus communication protocol to the existing lighting controller sensitivity to enable verification and remote control by external environmental factors, and also introduces a fuzzy inference system was configured by external environmental factors to control LED lighting. External environmental factors of temperature, humidity, illuminance value represented by the LED through a fuzzy control algorithm, the values accepted by the controller through the sensor. Modbus is using the RS485 Serial communication with other devices connected to the temperature, humidity, illumination and LED output status check is possible. In addition, the remote user is changed to enable it is possible to change the RGB values in the desired color change. Produced was confirmed that the LED controller output is based on the temperature, humidity and illumination.

• Journal of Navigation and Port Research
• /
• v.41 no.6
• /
• pp.395-400
• /
• 2017

As a serial communications protocol, Modbus has become a practically standard communication protocol and is now a commonly available means of connecting industrial electronic devices. Therefore, all devices can be connected using the Modbus protocol with the measurement and remote control on ships, buildings, trains, airplanes and more. The existing Modbus that has been used is based on serial communication. Modbus TCP uses Ethernet communication based on TCP / IP which is the most widely used Internet protocol today; so, it is faster than serial communication and can be connected to the Internet of Things. In this paper, we designed an algorithm to control LED lighting in a wireless Wi-Fi environment using the Modbus TCP communication protocol, and designed and implemented a LED controller circuit that can check external environmental factors and control remotely through the integrated management system of a ship. Temperature, humidity, current and illuminance values, which are external environmental factors, are received by the controller through the sensors, and these values are communicated to the ship's integrated management system via the Modbus protocol. The Modbus can be connected to Master devices via TCP communication to monitor temperature, humidity, current, illuminance status and LED output values, and also users can change the RGB value remotely in order to change to the desired color. In addition, in order to confirm the implementation of the controller, we developed a simulated ship management system to monitor the temperature, humidity, current and illumination conditions, and change the LED color of the controller by changing the RGB value remotely.