• Journal of Convergence Society for SMB
• /
• v.6 no.1
• /
• pp.33-38
• /
• 2016

Disease of patient who visited the hospital can cause different symptoms of the disease, depending on the environment and lifestyle. Recent medical services offered in patients has changed in the environment that can be selected for treatment by analyzing the patient according to the disease symptoms. In this paper, we propose an efficient method to manage disease control because the treatment method may change at any patients suffering from the disease according to the patient conditions by grouping the different treatments to patients for disease information. The proposed scheme has a feature that can be ingested by the patient big disease information, as well as to improve the treatment efficiency of the medical treatment the increase patient satisfaction. The proposed sheme can handle big data by clustering of disease information for patients suffering from diseases such as patient consent small groups. In addition, the proposed scheme has the advantage that can be conveniently accessed via a particular keyword, the treatment method according to patient disease information. The experimental results, the proposed method has been improved by 23% in terms of efficiency compared to conventional techniques, disease management time is gained 11.3% improved results. Medical service user satisfaction seen from the survey is to obtain a high 31.5% results.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.5
• /
• pp.461-467
• /
• 2020

In noisy and multi-speaker environments, the performance of speech recognition is unavoidably lower than in a clean environment. To improve speech recognition, in this paper, the signal of the speaker of interest is extracted from the mixed speech signals with multiple speakers. The VoiceFilter model is used to effectively separate overlapped speech signals. In this work, clustering by Probabilistic Linear Discriminant Analysis (PLDA) similarity score was employed to detect the speech signal of the interested speaker, which is used as the reference speaker to VoiceFilter-based separation. Therefore, by utilizing the speaker feature extracted from the detected speech by the proposed clustering method, this paper propose a speaker diarization system using only the mixed speech without an explicit reference speaker signal. We use phone-dataset consisting of two speakers to evaluate the performance of the speaker diarization system. Source to Distortion Ratio (SDR) of the operator (Rx) speech and customer speech (Tx) are 5.22 dB and -5.22 dB respectively before separation, and the results of the proposed separation system show 11.26 dB and 8.53 dB respectively.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2013.10a
• /
• pp.146-149
• /
• 2013

For in recently years, global cellular network service is changing rapidly to 4G. However, the fast introducing of 4G has been going with not enough research about security threat, it could be many kind of vulnerability. Therefore the research about security threat on 4G network is ongoing in external countries, but not sufficient in domestic. particularly in domestic situation of rapidly increased subscribers, The security threats which are hindering stability and usability could make a fatal effect on many users. 4G network should be considered about the feature of mobile network to protect 4G network stable. Mobile network has limited radio resources, it releases the radio resource which is not used in selected time and reallocate when detected the data transmission. Many signaling messages are transferred in the network entities to allocate or release the radio resource. In this paper, it will be introduced the technology to detect signaling DoS traffic hindering the stability and usability of network entities managing the radio resources by huge signaling message from the repetitive wireless connection/release message.

• Journal of Information Management
• /
• v.34 no.4
• /
• pp.63-84
• /
• 2003

This research is a case study that describes the NUCLIS21(Nuclear Information System 21), an integrated web-based information management system of KAERI(Korea Atomic Energy Research Institute), implemented to carry out the role of a national nuclear information center with government support as an information infra implementation programme. Through its user-centered single interface, the system aims at building an infrastructure for the national nuclear information center, as well as improving the information management system of the TID(Technical Information Department) within the institute. The system consists of two major parts which are an integrated module of the MIS and six different kinds of system. These include the Integral search system with OPAC, My Library, the Acquisition system, the Catalogue system, the Information Supply system, and the Serial Publication system. The DB is composed of Bibliographical DB, Original text DB and Abstract DB. A special feature of this system was designed as a unified network system through connection to MIS(Management Integration System) of the institute, so users can get research information for projects. Therefore, they have access to available information easily and access to the ongoing service of this system. Furthermore, users can share information by using our system. The survey has it that 75.7%(200 persons), 62.1%(164 persons) and 78.4%(207 persons) of the respondents are satisfied with the fidelity, speediness, and convenience of the system respectively, and the overall satisfaction of the respondents is comparably high.

• The Journal of the Convergence on Culture Technology
• /
• v.6 no.2
• /
• pp.509-514
• /
• 2020

The conventional TTS system consists of several modules, including text preprocessing, parsing analysis, grapheme-to-phoneme conversion, boundary analysis, prosody control, acoustic feature generation by acoustic model, and synthesized speech generation. But TTS system with deep learning is composed of Text2Mel process that generates spectrogram from text, and vocoder that synthesizes speech signals from spectrogram. In this paper, for the optimal Korean TTS system construction we apply Tacotron2 to Tex2Mel process, and as a vocoder we introduce the methods such as WaveNet, WaveRNN, and WaveGlow, and implement them to verify and compare their performance. Experimental results show that WaveNet has the highest MOS and the trained model is hundreds of megabytes in size, but the synthesis time is about 50 times the real time. WaveRNN shows MOS performance similar to that of WaveNet and the model size is several tens of megabytes, but this method also cannot be processed in real time. WaveGlow can handle real-time processing, but the model is several GB in size and MOS is the worst of the three vocoders. From the results of this study, the reference criteria for selecting the appropriate method according to the hardware environment in the field of applying the TTS system are presented in this paper.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.43 no.4 s.310
• /
• pp.74-86
• /
• 2006

This paper proposes a robust and computationally efficient algorithm for automatic video object segmentation. For implementing the spatio-temporal segmentation, which aims for efficient combination of the motion segmentation and the color segmentation, an SOM-based hierarchical clustering method in which the segmentation process is regarded as clustering of feature vectors is employed. As results, problems of high computational complexity which required for obtaining exact segmentation results in conventional video object segmentation methods, and the performance degradation due to noise are significantly reduced. A measure of motion vector reliability which employs MRF-based MAP estimation scheme has been introduced to minimize the influence from the motion estimation error. In addition, a noise elimination scheme based on the motion reliability histogram and a clustering validity index for automatically identifying the number of objects in the scene have been applied. A cross projection method for effective object tracking and a dynamic memory to maintain temporal coherency have been introduced as well. A set of experiments has been conducted over several video sequences to evaluate the proposed algorithm, and the efficiency in terms of computational complexity, robustness from noise, and higher segmentation accuracy of the proposed algorithm have been proved.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.48 no.1
• /
• pp.47-54
• /
• 2011

The remote control mobile robot is the robot accomplishing a task according to the orders giving by a user through departed communication system using a joystick. Basically, to supply a lot of information, as this type of robot uses visual information, the user can check the transmitted information by eyes and give orders to the robot. But the weak point of this type of robot is that it has a possibility to come into a collision with an obstacle not be seen to the user because of the communication delay occurring in a communication system and dead zone happening in visual information. To solve the problem, in this paper, we try to suggest a system applying a fuzzy control system to the robot to avoid collision with an obstacle by an immediate order of the user. The fuzzy control system has better performance than any other existing control methods in the change of noise and parameter. And it is more efficient than any other since it solves easy the complexity of the system analysis occurring because of the nonlinear feature of the mobile robot system. In this paper, we made experiments how the mobile robot controlled by the fuzzy control system avoids an obstacle, tracks the path and avoids the obstacle in the path, to prove the performance and to check the evaluation and the application possibility of the fuzzy control system.

• Journal of the Korean GEO-environmental Society
• /
• v.15 no.6
• /
• pp.67-73
• /
• 2014

Landslide disasters including debris flows are the one of the most frequent natural disasters in Korea, and losses of lives and property damages due to these catastrophic events have been increased every year. Various mitigation programs and related policies have been conducted in order to respond and prepare landslide disasters. Most landslide reduction programs are, however, focused on recovery actions after the disasters and lead to unrealistic consequences to the affected people and their properties. The main objective of this study, therefore, is to evaluate the landslide hazard based on the identification of geomorphological features, which is for the preparedness of the landslide disasters. Two methodologies, SINMAP and vector dispersion analyses are used to simulate those characteristics where landslides are actually located. Results showed that both methods well discriminate geomorphic features between stable and unstable domains. This proves that geomorphological characteristics well describe a relationship with the existing landslide hazard. SINMAP analysis which is based on the consecutive model considering external factors like infiltration is well identify the landslide hazard especially for debris flow type landslides rather than vector dispersion focusing on a specific area. Combining with other methods focusing specific characteristics of geomorphological feature, accurate landslide hazard assessments are implemented.

• Journal of IKEEE
• /
• v.9 no.1 s.16
• /
• pp.7-18
• /
• 2005

A key component of an autonomous mobile robot is to localize itself accurately and build a map of the environment simultaneously. In this paper, we propose a vision-based mobile robot localization and mapping algorithm using scale-invariant features. A camera with fisheye lens facing toward to ceiling is attached to the robot to acquire high-level features with scale invariance. These features are used in map building and localization process. As pre-processing, input images from fisheye lens are calibrated to remove radial distortion then labeling and convex hull techniques are used to segment ceiling region from wall region. At initial map building process, features are calculated for segmented regions and stored in map database. Features are continuously calculated from sequential input images and matched against existing map until map building process is finished. If features are not matched, they are added to the existing map. Localization is done simultaneously with feature matching at map building process. Localization. is performed when features are matched with existing map and map building database is updated at same time. The proposed method can perform a map building in 2 minutes on $50m^2$ area. The positioning accuracy is ${\pm}13cm$, the average error on robot angle with the positioning is ${\pm}3$ degree.

• Journal of Intelligence and Information Systems
• /
• v.18 no.3
• /
• pp.1-12
• /
• 2012

There are many trees in a roadside, parks or facilities for landscape. Although we are easily seeing a tree in around, it would be difficult to classify it and to get some information about it, such as its name, species and surroundings of the tree. To find them, you have to find the illustrated books for plants or search for them on internet. The important components of a tree are leaf, flower, bark, and so on. Generally we can classify the tree by its leaves. A leaf has the inherited features of the shape, vein, and so on. The shape is important role to decide what the tree is. And texture included in vein is also efficient feature to classify them. This paper evaluates the performance of a leaf classification system using both shape and texture features. We use Fourier descriptors for shape features, and both gray-level co-occurrence matrices and wavelets for texture features, and used combinations of such features for evaluation of images from the Flavia dataset. We compared the recognition rates and the precision-recall performances of these features. Various experiments showed that a combination of shape and texture gave better results for performance. The best came from the case of a combination of features of shape and texture with a flipped contour for a Fourier descriptor.