• Phonetics and Speech Sciences
• /
• v.1 no.3
• /
• pp.109-115
• /
• 2009

Many isolated word recognition systems may generate non-similar words for recognition candidates because they use only acoustic information. In this paper, we investigate several techniques which can exclude non-similar words from N-best candidate words by applying Levenstein distance measure. At first, word distance method based on phone and syllable distances are considered. These methods use just Levenstein distance on phones or double Levenstein distance algorithm on syllables of candidates. Next, word similarity approaches are presented that they use characters' position information of word candidates. Each character's position is labeled to inserted, deleted, and correct position after alignment between source and target string. The word similarities are obtained from characters' positional probabilities which mean the frequency ratio of the same characters' observations on the position. From experimental results, we can find that the proposed methods are effective for removing non-similar words without loss of system performance from the N-best recognition candidates of the systems.

• Molecules and Cells
• /
• v.43 no.7
• /
• pp.591-599
• /
• 2020

Complex cell-to-cell communication underlies the basic processes essential for homeostasis in the given tissue architecture. Obtaining quantitative gene-expression of cells in their native context has significantly advanced through single-cell RNA sequencing technologies along with mechanical and enzymatic tissue manipulation. This approach, however, is largely reliant on the physical dissociation of individual cells from the tissue, thus, resulting in a library with unaccounted positional information. To overcome this, positional information can be obtained by integrating imaging and positional barcoding. Collectively, spatial transcriptomics strategies provide tissue architecture-dependent as well as position-dependent cellular functions. This review discusses the current technologies for spatial transcriptomics ranging from the methods combining mechanical dissociation and single-cell RNA sequencing to computational spatial re-mapping.

• Spatial Information Research
• /
• v.7 no.1
• /
• pp.119-131
• /
• 1999

Various digital topographic maps including 1:1,000 and 1:5,000 have been produced for years and these digital maps are expected to serve a wide range of purposes. Most previous research about measuring positional accuracy for digital topographic maps depends int a method that acquires sample points from the tested map and compares to the corresponding known true locations using some statistics. Recently, a simple method for measuring positional accuracy for linear features is introduced but it has substantial problems applying to the spatial features such as roads, railroads, streams represented with double lines on the recently produced 1:1,000 and 1:5,000 digital topographic maps. This study develops and implements an effective method that measures road objects represented in the 1:5,000 digital topographic maps using a conventional GIS software. We also apply this method to a small study area located in Kangnam Gu, Seoul and illustrate the usefulness of this method.

• Journal of KIISE:Software and Applications
• /
• v.36 no.4
• /
• pp.335-345
• /
• 2009

In this paper, we propose a new web image caption extraction method considering the positional relation between a caption and an image and the lexical similarity between a caption and the main text containing the caption. The positional relation between a caption and an image represents how the caption is located with respect to the distance and the direction of the corresponding image. The lexical similarity between a caption and the main text indicates how likely the main text generates the caption of the image. Compared with previous image caption extraction approaches which only utilize the independent features of image and captions, the proposed approach can improve caption extraction recall rate, precision rate and 28% F-measure by including additional features of positional relation and lexical similarity.

• Journal of Korean Society for Geospatial Information Science
• /
• v.17 no.1
• /
• pp.97-104
• /
• 2009

Supplementary control point and urban control point have different purpose and characteristic in the view of installation but their measurement method and permanent marker position are very similar. Therefore if we could use them by connecting those two control point, we would improve the arrangement density as well as minimizing cost of restoration from the loss of control point. So we tried to suggest the possibility of connecting these two control point and usage by analysing positional accuracy of urban control point on the base of supplementary control point in this study. The result of this study is as following. When analysing the positional accuracy of 13 urban control points of the study areas, there were RMSE of connection errors between ${\pm}8cm$ and ${\pm}11cm$ for each measurement point. The result confirmed the possibility of connecting supplementary control point to urban control point and its usage within the allowable error tolerance that the present cadastral law permits.

• Journal of Korea Water Resources Association
• /
• v.55 no.spc1
• /
• pp.1187-1195
• /
• 2022

To run the algorithm for identifying the segments of a pipe network, the relative positional relation between nodes, pipes, and valves should be prepared as input information of a segment search algorithm. In order to more accurately identify the segments of real pipe network, pipe network GIS/CAD database that contains all isolation valves is more suitable than modeled pipe network information used for a hydraulic analysis program. In this study, we developed an algorithm that can establish the relative positional relations among node-pipe-valve suitable for pipe network segment search algorithms using GIS/CAD data of a real water supply network, and developed a MATLAB program that can implement it. The effectiveness of the developed MATLAB program was confirmed by applying it to a portion of a real municipal pipe network.

• Journal of Sensor Science and Technology
• /
• v.17 no.1
• /
• pp.53-60
• /
• 2008

Localization is the most important functions in mobile robots. There are so many approaches to realize this essential function in wheel based mobile robots, but it is not easy to find similar examples in small underwater robots. It is presented the sonar localization system using ubiquitous sensor network for a fish robot in this paper. A fish robot uses GPS and sonar system to find exact localization. Although GPS is essential tool to obtain positional information, this device doesn't provide reasonable resolution in localization. To obtain more precise localization information, we use several Ubiquitous Sensor Networks (USN) motes with sonar system. Experimental results show that a fish robot obtains more detailed positional information.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.692-695
• /
• 2024

With the recent advancements in artificial intelligence (AI), the performance of deep learning-based audio deepfake technology has significantly improved. This technology has been exploited for criminal activities, leading to various cases of victimization. To prevent such illicit outcomes, this paper proposes a deep learning-based audio deepfake detection model. In this study, we propose CoNSIST, an improved audio deepfake detection model, which incorporates three additional components into the graph-based end-to-end model AASIST: (i) Squeeze and Excitation, (ii) Positional Encoding, and (iii) Reformulated HS-GAL, This incorporation is expected to enable more effective feature extraction, elimination of unnecessary operations, and consideration of more diverse information, thereby improving the performance of the original AASIST. The results of multiple experiments indicate that CoNSIST has enhanced the performance of audio deepfake detection compared to existing models.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.11
• /
• pp.73-81
• /
• 2012

This paper proposes an ETWAD(Encapsulation and Tunneling Wormhole Attack Detection) design based on positional information and hop count on Ad-Hoc Network. The ETWAD technique is designed for generating GAK(Group Authentication Key) to ascertain the node ID and group key within Ad-hoc Network and authenticating a member of Ad-hoc Network by appending it to RREQ and RREP. In addition, A GeoWAD algorithm detecting Encapsulation and Tunneling Wormhole Attack by using a hop count about the number of Hops within RREP message and a critical value about the distance between a source node S and a destination node D is also presented in ETWAD technique. Therefore, as this paper is estimated as the average probability of Wormhole Attack detection 91%and average FPR 4.4%, it improves the reliability and probability of Wormhole Attack Detection.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.53-56
• /
• 2003

For identifying how a sample knit fabric is woven a method to obtain positional information of each yarn of the sample from voxel data made out of its x-ray CT images is newly proposed in this paper. The positional information is obtained by tracing the each yarn. The each yarn is traced by estimating a direction of the yarn in a certain small region in which the yarn can be regarded as straight and moving the region slightly along the estimated direction alternately. The yarn direction is estimated by correlating the voxel data in the region with a three-dimensional yarn model. The effectiveness of this method is confirmed by applying the method to voxel data made out of CT images of a knit fabric experimentally.