• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.19 no.5
• /
• pp.178-188
• /
• 2020

This paper proposes a lane recognition CNN network using split-attention network as a backbone to extract feature. Split-attention is a method of assigning weight to each channel of a feature map in the CNN feature extraction process; it can reliably extract the features of an image during the rapidly changing driving environment of a vehicle. The proposed deep neural networks in this paper were trained and evaluated using the Tusimple data set. The change in performance according to the number of layers of the backbone network was compared and analyzed. A result comparable to the latest research was obtained with an accuracy of up to 96.26, and FN showed the best result. Therefore, even in the driving environment of an actual vehicle, stable lane recognition is possible without misrecognition using the model proposed in this study.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.3
• /
• pp.155-164
• /
• 2009

Digital Content Marketplace based on Interactive Media is defmed as the marketplace for content service between contents supplier and consumer through iDTV environment. This Marketplace is increasing interest to u-Life service with Digital Environment. To Interactive Media, it can contribute to enhance its effectiveness by developing various contents and service model in the initial phase of broadcasting-communication convergence. This study designed metadata using Digital Content marketplace based on Interactive Media. Specially the matadata designing include recommendation-tag for supply supplementary content. It can support self-directed action. Through basic metadata with weight value, it is designed to support supplementary content customer to want on the marketplace. Recommendation-System can be built by many method and to recommend the service content including explicit properties using collaborative filtering method can solve limitations in existing content recommendation.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.15 no.6
• /
• pp.267-273
• /
• 2015

For newly incoming technologies owing to the advancements in information and communications technology, the new form of information occurs due to the combination form of personal information. In turn, personal information which is combination with existing personal information is on the increase. It is difficult to equalize the method in order to analysis the degree of risk for personal information because it is qualitative method which is defined on the current Personal Information Protection Law. This dissertation presents the model to assess the degree of risk by using the IPA(Importance-Performance Analysis) after measuring the importance and the weighted value for the personal information based on the existing the method of risk assessment. Through the model suggested in this dissertation, the subjective judgement can be excluded, the combination of personal information can be assessed and the standard criteria which is used as the objective indicators from the quantitative degree of risk can be suggested.

• Journal of Digital Convergence
• /
• v.12 no.5
• /
• pp.395-401
• /
• 2014

As social network service becomes common, the consumers use many discount coupons with which they can purchase goods via social commerce. Although, the quantities of coupons offered from social commerce are currently on the sharp increase, customized coupon service that reflects user preference is not offered. This paper proposes a coupon service method reflecting user's subjective inclination targeting food coupons to offer customized coupon service for social commerce. Towards this end, this paper conducts hierarchization of the factors that become standard in selecting coupons including food types, food prices, discount rates and the number of buyers. And then, this study classifies, extracts and offers the coupons using Fuzzy-AHP, a decision making support method that reflects subjective inclination. From the user satisfaction results on the extracted coupons, the users are generally satisfied: very satisfactory with 45%, satisfactory with 33% and fair with 22%, and there was no experiment participant, who was dissatisfied.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.47 no.9
• /
• pp.19-28
• /
• 2010

The GTS(Guaranteed Time Slot) of the IEEE 802.15.4 standard, which is the contention free access mechanism, is used for low-latency applications or applications requiring specific data bandwidth. But it has some problems such as delay of service due to FIFS(First In First Service) scheduling. In this paper, we proposes a weight based GTS allocation scheme for fair queuing in IEEE 802.15.4 LR-WPAN. The proposed scheme uses a weight that formed by how much more weight we give to the recent history than to the older history for a new GTS allocation. This scheme reduces service delay time and also guarantees transmission simultaneously within a limited time. The results of the performance analysis shows that our approach improves the performance as compared to the native explicit allocation mechanism defined in the IEEE 802.15.4 standard.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.26 no.4
• /
• pp.301-307
• /
• 2016

This paper proposes a lyric search method to consider the characteristics of the user query. According to the fact that queries for the lyric search are derived from highlight parts of the music, this paper uses the hierarchical agglomerative clustering to find the highlight and proposes a Gaussian weighting to consider the neighbor of the highlight as well as highlight. By setting the mean of a Gaussian weighting at the highlight, this weighting function has higher weights near the highlight and the lower weights far from the highlight. Then, this paper constructs a index of lyrics with the gaussian weighting. According to the experimental results on a data set obtained from 5 real users, the proposed method is proved to be effective.

• Journal of Korean Society of Forest Science
• /
• v.94 no.6
• /
• pp.387-396
• /
• 2005

Having employed the transitional probability model based on Markov chain, the study was carried out to examine successional trends for community types in the natural deciduous forest of Mt. Jumbong. The species composition of oncoming generation in overstory was estimated from that of mid-story, and the species composition in mid-story was based upon that of understory. Successional trend for each community was predicted from the reorganized probability matrix of tree replacement by the square of climax index, which was evaluated by the factors of light absorption, reproduction, and wood quality. As the result of analysis, following table shows the oncoming generation of steady state and dominant species in overstory and mid-story by community types. Even though Acer pseudo-sieboldianum and Carpinus cordata could hardly reach the canopy layer due to the intrinsic growth form, these species were predicted to maintain high compositional ratio so as to play an important ecological role in the study forest ecosystem.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.26 no.6
• /
• pp.1493-1504
• /
• 2016

As cyber threats have been increased over the Internet, the invasions of personal information are constantly occurring. A malicious user can access the Web site as a normal user using leaked personal information and does illegal activities. This paper proposes an effective method which authenticates a genuine user with real-time. The method use the user's profile which is a record of user's behavior created by Membership Analysis(MA) and Markov Chain Model(MCM). In addition to, user's profile is augmented by a Time Weight(TW) which reflects the user's tendency. This method can detect a malicious user who camouflage normal user. Even if it is a genuine user, it can be determined as an abnomal user if the user acts beyond the record profile. The result of experiment showed a high accuracy, 96%, for the correct user.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.6
• /
• pp.1294-1301
• /
• 2014

In this paper, separate probabilistic distribution models for voiced and unvoiced speech are estimated and utilized to improve speaker recognition performance. Also, in addition to the conventional mel-frequency cepstral coefficient, skewness, kurtosis, and harmonic-to-noise ratio are extracted and used for voiced speech intervals. Two kinds of scores for voiced and unvoiced speech are linearly fused with the optimal weight found by exhaustive search. The performance of the proposed speaker recognizer is compared with that of the conventional recognizer which uses mel-frequency cepstral coefficient and a unified probabilistic distribution function based on the Gassian mixture model. Experimental results show that the lower the number of Gaussian mixture, the greater the performance improvement by the proposed algorithm.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.5
• /
• pp.526-529
• /
• 2003

Recommender systems or collaborative filtering are methods to identify potentially interesting or valuable items to a particular user Under the assumption that people with similar interest tend to like the similar types of items, these methods use a database on the preference of a set of users and predict the rating on the items that the user has not rated. Usually the preference of a particular user is liable to vary with time and this temporal variation may cause an inaccurate identification and prediction. In this paper we propose a method to adapt the temporal variation of the user preference in order to improve the predictive performance of a collaborative filtering algorithm. To be more specific, the correlation weight of the GroupLens system which is a general formulation of statistical collaborative filtering algorithm is modified to reflect only recent similarity between two user. The proposed method is evaluated for EachMovie dataset and shows much better prediction results compared with GrouPLens system.