• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.1-23
• /
• 2018

From the 21st century, various high-quality services have come up with the growth of the internet or 'Information and Communication Technologies'. Especially, the scale of E-commerce industry in which Amazon and E-bay are standing out is exploding in a large way. As E-commerce grows, Customers could get what they want to buy easily while comparing various products because more products have been registered at online shopping malls. However, a problem has arisen with the growth of E-commerce. As too many products have been registered, it has become difficult for customers to search what they really need in the flood of products. When customers search for desired products with a generalized keyword, too many products have come out as a result. On the contrary, few products have been searched if customers type in details of products because concrete product-attributes have been registered rarely. In this situation, recognizing texts in images automatically with a machine can be a solution. Because bulk of product details are written in catalogs as image format, most of product information are not searched with text inputs in the current text-based searching system. It means if information in images can be converted to text format, customers can search products with product-details, which make them shop more conveniently. There are various existing OCR(Optical Character Recognition) programs which can recognize texts in images. But existing OCR programs are hard to be applied to catalog because they have problems in recognizing texts in certain circumstances, like texts are not big enough or fonts are not consistent. Therefore, this research suggests the way to recognize keywords in catalog with the Deep Learning algorithm which is state of the art in image-recognition area from 2010s. Single Shot Multibox Detector(SSD), which is a credited model for object-detection performance, can be used with structures re-designed to take into account the difference of text from object. But there is an issue that SSD model needs a lot of labeled-train data to be trained, because of the characteristic of deep learning algorithms, that it should be trained by supervised-learning. To collect data, we can try labelling location and classification information to texts in catalog manually. But if data are collected manually, many problems would come up. Some keywords would be missed because human can make mistakes while labelling train data. And it becomes too time-consuming to collect train data considering the scale of data needed or costly if a lot of workers are hired to shorten the time. Furthermore, if some specific keywords are needed to be trained, searching images that have the words would be difficult, as well. To solve the data issue, this research developed a program which create train data automatically. This program can make images which have various keywords and pictures like catalog and save location-information of keywords at the same time. With this program, not only data can be collected efficiently, but also the performance of SSD model becomes better. The SSD model recorded 81.99% of recognition rate with 20,000 data created by the program. Moreover, this research had an efficiency test of SSD model according to data differences to analyze what feature of data exert influence upon the performance of recognizing texts in images. As a result, it is figured out that the number of labeled keywords, the addition of overlapped keyword label, the existence of keywords that is not labeled, the spaces among keywords and the differences of background images are related to the performance of SSD model. This test can lead performance improvement of SSD model or other text-recognizing machine based on deep learning algorithm with high-quality data. SSD model which is re-designed to recognize texts in images and the program developed for creating train data are expected to contribute to improvement of searching system in E-commerce. Suppliers can put less time to register keywords for products and customers can search products with product-details which is written on the catalog.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.37 no.2
• /
• pp.31-43
• /
• 2000

This paper presents a novel algorithm lot extraction of the eye and mouth fields (facial features) from 2D gray level face images. First of all, it has been found that Eigenfeatures, derived from the eigenvalues and the eigenvectors of the binary edge data set constructed from the eye and mouth fields are very good features to locate these fields. The Eigenfeatures, extracted from the positive and negative training samples for the facial features, ate used to train a MultiLayer Perceptron(MLP) whose output indicates the degree to which a particular image window contains the eye or the mouth within itself. Second, to ensure robustness, the ensemble network consisting of multiple MLPs is used instead of a single MLP. The output of the ensemble network becomes the average of the multiple locations of the field each found by the constituent MLPs. Finally, in order to reduce the computation time, we extracted the coarse search region lot eyes and mouth by using prior information on face images. The advantages of the proposed approach includes that only a small number of frontal faces are sufficient to train the nets and furthermore, lends themselves to good generalization to non-frontal poses and even to other people's faces. It was also experimentally verified that the proposed algorithm is robust against slight variations of facial size and pose due to the generalization characteristics of neural networks.

• Journal of the Korean Society for Railway
• /
• v.16 no.4
• /
• pp.278-285
• /
• 2013

The suppression of residual settlement is required on earthwork sections as concrete track is introduced. Use of pile-slab structure is one of the settlement restraining methods applied on soft ground. The slab distributes the upper embankment load and piles transfer the load from the slab to the stiff ground. While this method is very effective in terms of load transfer, it has not yet been established for dealing with the vibration transfer effects and interaction characteristics between a structure and the ground. It is possible that vibration caused by a moving train load is propagated in the upper embankment, because the slab acts as a reflection layer and waves are multi-reflected. In this present paper, wave propagation generated by a moving train load is evaluated in the time and frequency domains to consider a roadbed structure using an artificial impact load and field measured train load. The results confirmed the wave reflection effect on the pile-slab structure, if the embankment height is sufficient, vibration propagation can be stably restrained, whereas if the height is not sufficient, the vibration amplitude is increased.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.26 no.1
• /
• pp.93-107
• /
• 2016

Game bot playing is one of the main risks in Massively Multi-Online Role Playing Games(MMORPG) because it damages overall game playing environment, especially the balance of the in-game economy. There have been many studies to detect game bot. However, the previous detection models require continuous maintenance efforts to train and learn the game bots' patterns whenever the game contents change. In this work, we have proposed a machine learning technique using the self-similarity property that is an intrinsic attribute in game bots and automated maintenance system. We have tested our method and implemented a system to major three commercial games in South Korea. As a result, our proposed system can detect and classify game bots with high accuracy.

• Journal of the Korean Society of Systems Engineering
• /
• v.7 no.2
• /
• pp.7-11
• /
• 2011

A high-speed railway system represents a typical example of large-scale multi-disciplinary system, consisting of subsystems such as rolling-stock, electrical hardware, electronics, control, information, communication, civil technology etc. The system design and acquisition data of the large-scale system must be the subject under strict configuration control and management. Systems engineering technology development project for Korea next generation High-speed Electric Multiple Unit (HEMU) system in progress is a national large system development project that is not only a large-size and complex but also multi-disciplinary in nature. Therefore, all stakeholders must understand and share the functional and performance requirements of HEMU throughout its life-cycle phases. Also in the test and evaluation phase, all systems requirements must be verified. In 2011, the prototype train manufacturing will be completed. It will do test run on the commercial line and all systems requirements are verified until 2012. For the system verification, the test and evaluation process have to be established before the test trial run. Using a systems engineering tool, the system design database(SDD) with requirements traceability and development process management in the course of the development have to be established. This paper represents the test and evaluation process development based on the SEMP(Systems Engineering Management Plan) developed in the design stage. The test and evaluation process is refined and updated in comparison to the design stage one. The test and evaluation process consists of procedure, test and evaluation method and schedule. So through this process, it is defined that each systems requirements is verified on which test and about what time.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.5
• /
• pp.1044-1051
• /
• 2003

This paper surveys the EBP(Error Back Propagation) learning, the Cross Entropy function and the LBL(Layer By Layer) learning, which are used for learning the MLP(Multi Layer Perceptrons). We compare the merits and demerits of each learning method in the handwritten digit recognition. Although the speed of EBP learning is slower than other learning methods in the initial learning process, its generalization capability is better. Also, the speed of Cross Entropy function that makes up for the weak points of EBP learning is faster than that of EBP learning. But its generalization capability is worse because the error signal of the output layer trains the target vector linearly. The speed of LBL learning is the fastest speed among the other learning methods in the initial learning process. However, it can't train for more after a certain time, it has the lowest generalization capability. Therefore, this paper proposes the standard of selecting the learning method when we apply the MLP.

• Journal of IKEEE
• /
• v.12 no.1
• /
• pp.65-74
• /
• 2008

One of the popular methods used for pattern classification is the MBR (Memory-Based Reasoning) algorithm. Since it simply computes distances between a test pattern and training patterns or hyperplanes stored in memory, and then assigns the class of the nearest training pattern, it is notorious for memory usage and can't learn additional information from new data. In order to overcome this problem, we propose an incremental learning algorithm (iMPA). iMPA divides the entire pattern space into fixed number partitions, and generates representatives from each partition. Also, due to the fact that it can not learn additional information from new data, we present iMPA which can learn additional information from new data and not require access to the original data, used to train. Proposed methods have been successfully shown to exhibit comparable performance to k-NN with a lot less number of patterns and better result than EACH system which implements the NGE theory using benchmark data sets from UCI Machine Learning Repository.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.40 no.6
• /
• pp.90-97
• /
• 2003

This paper proposes a speech recognition based on HMM(Hidden Markov Model) using FSVQ(First Section Vector Quantization) and fuzzy concept. In the proposed paper, we generate codebook of First Section, and then obtain multi-observation sequences by order of large propabilistic values based on fuzzy rule from the codebook of the first section. Thereafter, this observation sequences of first section from codebooks is trained and in case of recognition, a word that has the most highest probability of first section is selected as a recognized word by same concept. Train station names are selected as the target recognition vocabulary and LPC cepstrum coefficients are used as the feature parameters. Besides the speech recognition experiments of proposed method, we experiment the other methods under same conditions and data. Through the experiment results, it is proved that the proposed method based on HMM using FSVQ and fuzzy concept is superior to tile others in recognition rate.

• Journal of KIISE:Software and Applications
• /
• v.33 no.10
• /
• pp.886-895
• /
• 2006

Fingerprint classification reduces the number of matches required in automated fingerprint identification systems by categorizing fingerprints into a predefined class. Support vector machines (SVMs), widely used in pattern classification, have produced a high accuracy rate when performing fingerprint classification. In order to effectively apply SVMs to multi-class fingerprint classification systems, we propose a novel method in which SVMs are generated with the one-vs-all (OVA) scheme and dynamically ordered with $na{\ddot{i}}ve$ Bayes classifiers. More specifically, it uses representative fingerprint features such as the FingerCode, singularities and pseudo ridges to train the OVA SVMs and $na{\ddot{i}}ve$ Bayes classifiers. The proposed method has been validated on the NIST-4 database and produced a classification accuracy of 90.8% for 5-class classification. Especially, it has effectively managed tie problems usually occurred in applying OVA SVMs to multi-class classification.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.61 no.1
• /
• pp.179-185
• /
• 2012

Since the eco-era is getting closer, the importance of noise reducing in the passenger cars of high-speed train is very important. The active noise control is best choice to reduce low frequency noise because the passive one is too heavy for high speed trains where weight is so critical. Also ANC is able to reduce the ambient noise when the environmental-factor changes. To reduce a three-dimensional closed-space sound field like a car of a high-speed rail is hard to do using single channel ANC control system. We used multi-channel FXLMS algorithm which calculation speed is fast and the secondary path estimation is possible in order to take into account the physical delay in electro acoustic hardware control loudspeaker and power amplifier. Firstly, we have measured interior noise of KTX and estimated noise path in KTX test-bed. However there was some problem related to algorithm divergence and increasing the filter order. We have made a simulation of interior environment of KTX car by using three frequency bands of 120Hz, 280Hz, 360Hz as the most important for KTX ANC system. During this research the interior noise reduction of KTX car was made by using the multi-channel FXLMS algorithm. Reduction performance was evaluated and compared each other for open space section and tunnel section. in-situ experiment for the KTX noise reduction by proposed ANC was performed based on data obtained in simulation and they were compared for open space section and tunnel section as well.