• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.29 no.3
• /
• pp.297-320
• /
• 2018

This study aims to improve the access and retrieval of images and to find a way to effectively generate tags as a tool for providing explanation of images. To do this, this study investigated the features of human tagging and machine tagging, and compare and analyze them. Machine tags had the highest general attributes, some specific attributes and visual elements, and few abstract attributes. The general attribute of the human tag was the highest, but the specific attribute was high for the object and scene where the human tag constructor can recognize the name. In addition, sentiments and emotions, as well as subjects of abstract concepts, events, places, time, and relationships are represented by various tags. The tag set generated through this study can be used as basic data for constructing training data set to improve the machine learning algorithm.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.1A
• /
• pp.50-57
• /
• 2009

Tag collection is one of the major functions in RFID systems. In this paper, we propose a multichannel slotted-aloha anticollision protocol to mitigate the tag collision problem and improve tag collection performance in active RFID systems. In the proposed anticollision protocol, while an RFID reader transmits commands to multiple RFID tags via a common channel, it receives multiple tag responses over multiple channels simultaneously. We also implemented an active RFID system supporting the proposed anticollision protocol. In experiments with the implemented reader and 60 tags, the proposed dual-channel slotted-aloha anticollision protocol showed a large improvement in tag collection performance compared with the single-channel slotted-aloha anticollision protocol. With 60 tags, the average time spent on tag collection using the dual-channel anticollision protocol was 600.543 ms, which was 46.3% of the 1289.721 ms required using the single-channel anticollision protocol.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38B no.3
• /
• pp.199-206
• /
• 2013

This paper proposes a modified tag collection algorithm that improves the drawback of the identified slot scan-based tag collection algorithm presented in a previous paper to improve the tag collection performance in active RFID systems. The previous identified slot scan-based tag collection algorithm is optimized in situations where all the tags store the fixed size of data, so it could not result in a good performance improvement with tags having the variable size of data. The improved tag collection algorithm proposed in this paper first collects the slot size information required for the data transmission from each tag via the identified slot scan phase, and then performs the tag collection phase using the information, which resolves the problem of the previous identified slot scan-based tag collection algorithm. The simulation results for performance evaluation showed that the proposed tag collection algorithm resulted in the almost same tag collection performance as the previous algorithm when all the tags have the same size of data and led a large improvement of the tag collection performance in ISO/IEC 18000-7 unlike the previous algorithm when each tag has a random size of data.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.21 no.7
• /
• pp.752-760
• /
• 2010

This paper investigates the performance of EPCglobal Generation-2(Gen-2) radio frequency identification(RFID) protocol under reader-to-reader interfering environments. We establish a modified discrete-time Markov chain(DTMC) model for the Gen-2 and accordingly obtain quantitative results of query success rate(QSR) and tag identification speed (TIS). Extensive simulations validate our theoretical analysis and demonstrate that the number of tags over 100 has little impact on the performance. TIS linearly decreases by 10 tags/sec/reader as the number of interfering readers increases. Our model for Gen-2 protocol is also useful to study the performance of other RFID protocols.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.2
• /
• pp.459-462
• /
• 2005

In this paper, we investigate the high precision location estimation algorithm in 2.45GHz band RTLS with multiple tags. The location is estimated in LOS environments, 300m ${\times}$ 300m area, and 2D coordinates adopting a TDOA scheme which is not necessitate the transmission time of tags. We evaluate the average estimation error in distance assuming that tags are randomly distributed and the readers(3${\sim}$8) are uniformly(equal space) placed in test area. In results, average estimation error is 3.12m and 1.47m at reader numbers of 4 and 8, respectively. Minimum estimation error is obtained when the accumulated receiving signal from a tag is 3 or 4 regardless of available reader numbers. The error is less than 3m, satisfies the specification of RTLS.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.6B
• /
• pp.618-630
• /
• 2011

Mobile RFID system, that consists of the existing RFID reader mounted on the mobile devices such as smartphones, is able to provide the users a variety of services and convenience. Although the users can get the information about a certain product anytime anywhere, there is high probability that their privacy may be violated because their belongings with RFID tags can be scanned by other mobile readers at any time. Several RFID authentication schemes have been proposed to deal with these privacy issues. However, since the existing solutions require heavy computation on the tag side, most of them is not applicable to the general low-cost passive tags which do not have any processing unit. In this paper, we propose the efficient authentication scheme for mobile RFID system applicable to the passive tags as well as the active ones by the best use of computing capability of mobile devices. The proposed scheme satisfies the import security issues such as tag protection, untraceability, anti-traffic analysis. We also implement the proposed scheme on top of real smartphone for feasibility and show the experimental results from it.

• Environmental Analysis Health and Toxicology
• /
• v.26
• /
• pp.10.1-10.7
• /
• 2011

Objects: Chironomus riparius, a non-biting midge (Chironomidae, Diptera), is extensively used as a model organism in aquatic ecotoxicological studies, and considering the potential of C. riparius larvae as a bio-monitoring species, little is known about its genome sequences. This study reports the results of an Expressed Sequence Tags (ESTs) sequencing project conducted on C. riparius larvae using 454 pyrosequencing. Method: To gain a better understanding of C. riparius transcriptome, we generated ESTs database of C.ripairus using pyrosequencing method. Results: Sequencing runs, using normalized cDNA collections from fourth instar larvae, yielded 20,020 expressed sequence tags, which were assembled into 8,565 contigs and 11,455 singletons. Sequence analysis was performed by BlastX search against the National Center for Biotechnology Information (NCBI) nucleotide (nr) and uniprot protein database. Based on the gene ontology classifications, 24% (E-value${\leq}1^{-5}$) of the sequences had known gene functions, 24% had unknown functions and 52% of sequences did not match any known sequences in the existing database. Sequence comparison revealed 81% of the genes have homologous genes among other insects belonging to the order Diptera providing tools for comparative genome analyses. Targeted searches using these annotations identified genes associated with essential metabolic pathways, signaling pathways, detoxification of toxic metabolites and stress response genes of ecotoxicological interest. Conclusions: The results obtained from this study would eventually make ecotoxicogenomics possible in a truly environmentally relevant species, such as, C. riparius.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2006.05a
• /
• pp.931-934
• /
• 2006

Recently, as the problems related to agricultural product's safety occur continuously, the system with traceability that can quickly trace back the agricultural products with problems occurred and enable the analysis of causes and prohibition of distribution is required. Therefore, the recording of distribution data is important for this system and for this purpose, the RFID technology that can perceive tags from remote place through radio frequency can be effectively used. But, since the RFID is a technology that is being watched recently, the system that processes RFID tags and distribution data is trifling. Hereby, this thesis designed and implemented the RFID processing system that enables the career trace by analyzing and managing the data on recorded RFID tags and career data based on the standard of traceability.

• Journal of KIISE
• /
• v.43 no.1
• /
• pp.47-53
• /
• 2016

While the Internet develops rapidly, a huge amount of image data collected from smart phones, digital cameras and black boxes are being shared through social media sites. Generally, social images are handled by tagging them with information. Due to the ease of sharing multimedia and the explosive increase in the amount of tag information, it may be considered too much hassle by some users to put the tags on images. Image retrieval is likely to be less accurate when tags are absent or mislabeled. In this paper, we suggest a method of extracting tags from social images by using image content. In this method, CNN(Convolutional Neural Network) is trained using ImageNet images with labels in the training set, and it extracts labels from instagram images. We use the extracted labels for automatic image tagging. The experimental results show that the accuracy is higher than that of instagram retrievals.

• KIPS Transactions on Computer and Communication Systems
• /
• v.2 no.11
• /
• pp.483-488
• /
• 2013

Users can access their data anywhere, at any time by using outsourced storage. But they cannot know how service provider manage the data. Even user cannot know when data damaged. To solve these problems, the outsourced storage auditing schemes has been proposed. Most proposed schemes are based on Homomorphic Verifiable Tags. But it has computational efficiency limitation because data used to exponent. In this paper, we propose a novel approach to outsourced storage auditing scheme using coefficient matrix. In the proposed scheme, data used to auditing by coefficient matrix form. Auditing procedures are proceed as solving the linear simultaneous equation. The auditor can audit easily by solving the equation using solution vector. The auditor can audit the n size data using sqrt(n) size data through out proposed scheme.