• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.5
• /
• pp.669-674
• /
• 2018

Data augmentation is an efficient way to reduce overfitting on models and to improve a performance supplementing extra data for training. It is more important in deep learning based industrial machine vision. Because deep learning requires huge scale of learning data to learn a model, but acquisition of data can be limited in most of industrial applications. A very generic method for augmenting image data is to perform geometric transformations, such as cropping, rotating, translating and adjusting brightness of the image. The effectiveness of data augmentation in image classification has been reported, but it is rare in defect inspections. We explore and compare various basic augmenting operations for the metal surface defects. The experiments were executed for various types of defects and different CNN networks and analysed for performance improvements by the data augmentations.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.63 no.8
• /
• pp.1111-1115
• /
• 2014

With the increasing interest of a demand side management using a Smart Grid infrastructure, the demand resources and energy usage data management becomes an important factor in energy industry. In addition, with the help of Advanced Measuring Infrastructure(AMI), energy usage data becomes a Big Data System. Therefore, it becomes difficult to store and manage the demand resources big data using a traditional relational database management system. Furthermore, not many researches have been done to analyze the big energy data collected using AMI. In this paper, we are proposing a Hadoop based Big Data system to manage the demand resources energy data and we will also show how the demand side management systems can be used to improve energy efficiency.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.53 no.7
• /
• pp.525-530
• /
• 2004

A classifier was constructed by using a generalized regression neural network (GRU) and random generator (RG), which was applied to classify DNA sequences. Three data sets evaluated are eukaryotic and prokaryotic sequences (Data-I), eukaryotic sequences (Data-II), and prokaryotic sequences (Data-III). For each data set, the classifier performance was examined in terms of the total classification sensitivity (TCS), individual classification sensitivity (ICS), total prediction accuracy (TPA), and individual prediction accuracy (IPA). For a given spread, the RG played a role of generating a number of sets of spreads for gaussian functions in the pattern layer Compared to the GRNN, the RG-GRNN significantly improved the TCS by more than 50%, 60%, and 40% for Data-I, Data-II, and Data-III, respectively. The RG-GRNN also demonstrated improved TPA for all data types. In conclusion, the proposed RG-GRNN can effectively be used to classify a large, multivariable promoter sequences.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.12
• /
• pp.1751-1758
• /
• 2017

This paper presents a fuzzy logic system with data preprocessing to make the accurate electric power load prediction system. The fuzzy logic system acceptably treats the hidden characteristic of the nonlinear data. The data preprocessing processes the original data to provide more information of its characteristics. Thus the combination of two methods can predict the given data more accurately. The former uses TSK fuzzy logic system to apply the linguistic rule base and the linear regression model while the latter uses the linear interpolation method. Finally, four regional electric power load data in taiwan are used to evaluate the performance of the proposed prediction system.

• The KIPS Transactions:PartD
• /
• v.16D no.5
• /
• pp.673-682
• /
• 2009

XML data are widely used for data representation and exchange on the Web and the data type is an continuous stream in ubiquitous environment. Therefore there are some mining researches related to the extracting of frequent structures and the efficient query processing of XML stream data. In this paper, we propose a mining method to extract frequent structures of XML stream data in recent window based on the sliding window. XML stream data are modeled as a tree set, called XFP_tree and we quickly extract the frequent structures over recent XML data in the XFP_tree.

• Journal of Convergence for Information Technology
• /
• v.12 no.4
• /
• pp.103-110
• /
• 2022

Online transactions are more familiar in various fields due to the development of the ICT and the increase in trading platforms. In particular, the amount of transactions is increasing due to the increase in used transaction platforms and users, and reliability is very important due to the nature of used transactions. Among them, the used car market is very active because automobiles are operated over a long period of time. However, used car transactions are a representative market to which information asymmetry is applied. In this paper presents a DID-based transaction model that guarantees reliability to solve problems with false advertisements and false sales in used car transactions. In the used car transaction model, sellers only register data issued by the issuing agency to prevent false sales at the time of initial sales registration. It is authenticated with DID Auth in the issuance process, it is safe from attacks such as sniping and middleman attacks. In the presented transaction model, integrity is verified with VP's Proof item to increase reliability and solve information asymmetry. Also, through direct transactions between buyers and sellers, there is no third-party intervention, which has the effect of reducing fees.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.50 no.5
• /
• pp.233-240
• /
• 2001

This paper presents a time series prediction method using a fuzzy rule-based system. Extracting fuzzy rules by performing a simple one-pass operation on the training data is quite attractive because it is easy to understand, verify, and extend. The simplest method is probably to relate an estimate, x(n+k), with past data such as x(n), x(n-1), ..x(n-m), where k and m are prefixed positive integers. The relation is represented by fuzzy if-then rules, where the past data stand for premise part and the predicted value for consequence part. However, a serious problem of the method is that it cannot handle nonstationary data whose long-term mean is varying. To cope with this, a new training method is proposed, which utilizes the difference of consecutive data in a time series. In this paper, typical previous works relating time series prediction are briefly surveyed and a new method is proposed to overcome the difficulty of prediction nonstationary data. Finally, computer simulations are illustrated to show the improved results for various time series.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.26 no.3
• /
• pp.560-569
• /
• 2002

The segmentation has been performed to the data of good quality in most cases, so the adoption of previous segmentation theory to the measured data with a laser scanner does not produce good result because of the characteristics of the data with noise component. A new approach to perform the segmentation on the scanned data is introduced to deal with problems during reverse engineering process. A triangular net is generated from measured point data, and the segmentation on it is classified as plane, smooth and rough segment. The segmentation result in each segment depends on the user-defined criteria. And the difference of the segmentation between the data of good quality and the data with noise is described and analyzed with several real models. The segment boundaries selected are used to maintain the characteristics of the parts during modeling process, thus they contribute to the automation of the reverse engineering.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.18 no.1
• /
• pp.31-36
• /
• 2010

Collecting all failures during life cycle of vehicle is not easy way because its life cycle is normally over 10 years. Warranty period can help gathering failures data because most customers try to repair its failures during warranty period even though small failures. This warranty data, which means failures during warranty period, can be a good resource to predict initial reliability and permanence reliability. However uncertainty regarding reliability prediction remains because this data is censored. University of Wuppertal and major auto supplier developed the reliability prognosis model considering censored data and this model introduce to predict reliability estimate further "failure candidate". This paper predicts reliability of telecommunications system in vehicle using the model and describes data structure for reliability prediction.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.1
• /
• pp.436-453
• /
• 2018

Storage cloud scheme, pushing data to the storage cloud poses much attention regarding data confidentiality. With encryption concept, data accessibility is limited because of encrypted data. To secure storage system with high access power is complicated due to dispersed storage environment. In this paper, we propose a hardware-based security scheme such that a secure dispersed storage system using erasure code is articulated. We designed a hardware-based security scheme with data encoding operations and migration capabilities. Using TPM (Trusted Platform Module), the data integrity and security is evaluated and achieved.