• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.6
• /
• pp.175-182
• /
• 2020

The non-contact eye tracking is a nonintrusive human-computer interface providing hands-free communications for people with severe disabilities. Recently. it is expected to do an important role in non-contact systems due to the recent coronavirus COVID-19, etc. This paper proposes a novel approach for an eye mouse using an eye tracking method based on a context-aware based AdaBoost multi-region classifier and ASSL algorithm. The conventional AdaBoost algorithm, however, cannot provide sufficiently reliable performance in face tracking for eye cursor pointing estimation, because it cannot take advantage of the spatial context relations among facial features. Therefore, we propose the eye-region context based AdaBoost multiple classifier for the efficient non-contact gaze tracking and mouse implementation. The proposed method detects, tracks, and aggregates various eye features to evaluate the gaze and adjusts active and semi-supervised learning based on the on-screen cursor. The proposed system has been successfully employed in eye location, and it can also be used to detect and track eye features. This system controls the computer cursor along the user's gaze and it was postprocessing by applying Gaussian modeling to prevent shaking during the real-time tracking using Kalman filter. In this system, target objects were randomly generated and the eye tracking performance was analyzed according to the Fits law in real time. It is expected that the utilization of non-contact interfaces.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.2
• /
• pp.31-41
• /
• 2024

Deep learning models show excellent performance in tasks such as image classification and object detection in the field of computer vision, and are used in various ways in actual industrial sites. Recently, research on improving robustness has been actively conducted, along with pointing out that this deep learning model is vulnerable to hostile examples. A hostile example is an image in which small noise is added to induce misclassification, and can pose a significant threat when applying a deep learning model to a real environment. In this paper, we tried to confirm the robustness of the edge-learning classification model and the performance of the adversarial example detection model using it for adversarial examples of various algorithms. As a result of robustness experiments, the basic classification model showed about 17% accuracy for the FGSM algorithm, while the edge-learning models maintained accuracy in the 60-70% range, and the basic classification model showed accuracy in the 0-1% range for the PGD/DeepFool/CW algorithm, while the edge-learning models maintained accuracy in 80-90%. As a result of the adversarial example detection experiment, a high detection rate of 91-95% was confirmed for all algorithms of FGSM/PGD/DeepFool/CW. By presenting the possibility of defending against various hostile algorithms through this study, it is expected to improve the safety and reliability of deep learning models in various industries using computer vision.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.25 no.5
• /
• pp.391-399
• /
• 2005

Laser-generated ultrasonic inspection system is a non-contact scanning inspection device with high spatial resolution and wide bandwidth. The amplitude of laser-generated ultrasound is varied according to the energy of pulse laser and the surface conditions of an object where the CW measuring laser beam is pointing. In this paper, we correct the generating errors by measuring the energy of pulse laser beam and correct the measuring errors by extracting the gain information of laser interferometer at each time. h dynamic stabilizer is developed to stably scan on the surface of an object for an laser-generated ultrasonic inspection system. The developed system generates ultrasound after adaptively finding the maximum gain time of an laser interferometer and processes the signal in real time after digitization with high speed. In this paper, we describe hardware configuration and control algorithm to build a stable laser-generated ultrasonic inspection system. Also, we confirmed through experiments that the proposed correction method for the generating errors and measuring errors is effective to improve the performance of a system.

• Journal of KIISE:Information Networking
• /
• v.28 no.2
• /
• pp.262-276
• /
• 2001

Recently much research for developing forwarding table that support fast router without employing both special hardware and new protocols. This article introduces a new forwarding data structure based on the software to enable forwarding lookup to be penormed at giga-bit speed. The forwarding table is known as a bottleneck of the routers penormance due to its high complexity proportional to the forwarding table size. The recent research that based on the software uses a Patricia trie and its variants. and also uses a hash function with prefix length key and others. The proposed forwarding table structure construct a forwarding table by the bit stream array in which it constructs trie from routing table prefix entries and it represents each pointer pointing the child node and the associated forwarding table entry with one bit The trie structure and routing prefix pointer need a large memory when representing those by linked-list or array. but in the proposed data structure, the needed memory size is small enough since it represents information with one bit. Additionally, by use a lookup method that start searching at desired middle level we can shorten the search path. The introduced data structure. called bit-map trie shows that we can implement a fast forwarding engine on the conventional Pentium processor by reducing the backbone routing table fits into Level 2 cache of Pentium II processor and shortens the searching path. Our experiments to evaluate the performance of proposed method show that this bit-map trie accomplishes 5.7 million lookups per second.

• Korean Security Journal
• /
• no.44
• /
• pp.251-278
• /
• 2015

Registered security guards carry out police duties as civilian police who are in charge of security service, and so they have a two-fold status: a civilian in terms of a social standing and a policeman in the way that they execute the authority of security. The problem caused by this legal position is that their legal rights and obligation can be unclear in the task-action and working relationship. This paper attempts to study their functions, rights, and legal duties through the interpretation of the related positive law so as to reveal the problems that may spring from this ambiguous status of registered security guards. This endeavor illuminates their legal status specified in the positive law in and around the Act on the police assigned for special guard, observing their functions and the legal duties in the pursuit of their tasks, and ending up pointing out the problems of the positive law. As a result of research work, the most significant problems, even if multifarious, are the avoidance of the state reparation in the responsibility for the illegal behavior in connection with their operation; the unconstitutionality of the disciplinary punishment regulation originated from the entrust with full powers; the imperfection of the rules about the cooperative ties with the police; the possibility of human rights abuse caused by the ban on the labor dispute; the equality problems from the dual pay system; and the inadequacy of the codes about the recruitment qualification and method. This research is intended to help achieve the purpose of the security of national critical facilities through the smooth execute of duties as well as the protection of the guards' rights. Besides, the key focuses posed in this paper are worthy of being developed more accurately through the following researches.

• 한국HCI학회:학술대회논문집
• /
• 2008.02b
• /
• pp.47-52
• /
• 2008

J.E.N(Joint Enjoyment Network) is the ubiquitous UCC play contents system that provides amusements off-line by utilizing the UCC that people are able to participate in. It is also the entertainment service system that is devised to use contents that could be enjoyed through the cell phone and to make people participate in production, in cases of meeting people off-line during their pastime. The development of this system begins from pointing out the problem of the limitation of the UCC that does participate in production but the produced contents could only be watched, and the UCC only works on-line. The goal of researching this ubiquitous UCC play contents is to suggest the UCC entertainment service system that are available off-line by applying the interactive system to UCC and, by this, to obtain the positive responses about the off-line participating UCC contents. In this research, it chooses scenario and 테스크 performance as the method of deriving the participating UCC from the off-line, and it suggests the ubiquitous UCC play contents system through analyzing the behaviors in each off-line place. Moreover, by developing the prototype that could practically perform the system, it verifies the potentialities of the UCC entertainment service. The anticipation of this study is to make people participate and enjoy not only in production, but also in the contents made. This research will provide the modern people who concern about where to go and what to do with the amusements and will also present a way of new form of contents mixed with UCC and the entertainment service.

• Journal of The Korean Association For Science Education
• /
• v.22 no.3
• /
• pp.411-421
• /
• 2002

This study investigated small group processes in paired think-aloud problem solving. Two high school chemistry classes were assigned to St-SL group (using Strategy-Solve Listener) and SL group (Solver Listener), and their small-group behaviors were audio/video taped. Verbal behaviors of solver and listener in respect to 4 problem-solving stages and performance levels at each stage were analyzed. At the understanding stage, listeners in the St-SL group exhibited more behaviors of agreement to solver's understanding processes about given and goal of problem. As regards recalling a related law at the planning stage, solvers in the St-SL group exhibited more behaviors of modification based on listener's questions or pointing out. These verbal interactions seemed to have a positive effect on students' deriving the physical quantity with the proper laws. Few in both SL and St-SL groups exhibited the behaviors regarding setting up subgoals. No verbal behavior was observed in the SL group at the reviewing stage, and solvers in the St-SL group tended to ask for listener's agreement. However, only few performed the strategy explaining the meaning of answer at the molecular level correctly through the interactions. The St-SL group perceived that the understanding stage was the most helpful and that the planning or reviewing stages were difficult to apply.

• Smart Media Journal
• /
• v.8 no.2
• /
• pp.58-65
• /
• 2019

Automatic text summarization is a process of shortening a text document by either extraction or abstraction. The abstraction approach inspired by deep learning methods scaling to a large amount of document is applied in recent work. Abstractive text summarization involves utilizing pre-generated word embedding information. Low-frequent but salient words such as terminologies are seldom included to dictionaries, that are so called, out-of-vocabulary(OOV) problems. OOV deteriorates the performance of Encoder-Decoder model in neural network. In order to address OOV words in abstractive text summarization, we propose a copy mechanism to facilitate copying new words in the target document and generating summary sentences. Different from the previous studies, the proposed approach combines accurate pointing information and selective copy mechanism based on bidirectional RNN and bidirectional LSTM. In addition, neural network gate model to estimate the generation probability and the loss function to optimize the entire abstraction model has been applied. The dataset has been constructed from the collection of abstractions and titles of journal articles. Experimental results demonstrate that both ROUGE-1 (based on word recall) and ROUGE-L (employed longest common subsequence) of the proposed Encoding-Decoding model have been improved to 47.01 and 29.55, respectively.

• Smart Media Journal
• /
• v.10 no.1
• /
• pp.9-15
• /
• 2021

The eye tracking system designed in this paper is an eye-based computer input device designed to give an easy access for those who are uncomfortable with Lou Gehrig's or various muscle-related diseases. It is an eye-based-computer-using device for users whose potential demand alone amounts to 30,000. Combining the number of Lou Gehrig's patients in Korea estimated at around 1,700, and those who are unable to move their bodies due to various accidents or diseases. Because these eye input devices are intended for a small group of users, many types of commercial devices are available on the market. It is making them more expensive and difficult to use for these potential users, less accessible. For this reason, each individual's economic situation and individual experience with smart devices are slightly different. Therefore, making it difficult to access them in terms of cost or usability to use a commercial eye tracking system. Accordingly, attempts to improve accessibility to IT devices through low-cost but easy-to-use technologies are essential. Thus, this paper proposes a complementary superior performance eye tracking system that can be conveniently used by far more people and patients by improving the deficiencies of the existing system. Through voluntary VoCs(Voice of Customers) of users who have used different kinds of eye tracking systems that satisfies it through various usability tests, and we propose a reduced system that the amount of calculation to 1/15th, and eye-gaze tracking error rate to 0.5~1 degree under.

• (The) Research of the performance art and culture
• /
• no.36
• /
• pp.415-440
• /
• 2018

On September 23, 2015, the Ministry of Education announced the 2015 revision of educational curriculum which aimed at 'cultivating creative talents' based on the Article 23, Section 2 of the Elementary and Secondary Education Law. As a result, music curriculum have also been partially revised, which seems to maintain the 2009 revision of music curriculum. Although Jeongganbo 井間譜 is already exposed in the music curriculum for the third and forth grades of elementary school, the learning content about how to read Jeongganbo and how to express the pitch and length of sound including the origin of its name and the background of its invention are dealt with specifically in the fifth and sixth grades. Jeongganbo is known as the oldest mensural notation in the Orient created by King Sejong of the Joseon Dynasty in the middle of the $15^{th}$ century, and it was used for the first time in Sejong sillok akbo 世宗實錄樂譜 (Scores in the Annals of King Sejong), the oldest musical score still in existence. However, in the music textbooks as well as the most of specialized books related to the Korean traditional music, it is uncritically accepted without providing clear grounds that Sejong invented Jeongganbo himself. If so, it is necessary to investigate on which grounds it is claimed that Sejong invented Jeongganbo. This paper first examined the grounds of the proposition that "Sejong invented Jeongganbo," which is introduced in the music textbooks for the fifth and sixth grades of elementary school, by separating it into Sejong's creation of Sinak 新樂 (new music), Sejo's invention of Jeongganbo and Sejong's invention of Hangeul. Next, this paper examined how the subject of the invention of Jeongganbo has been described in the textbooks for the fifth and sixth grades in elementary school based on the 2009 revision of music curriculum, and suggested the direction of a desirable music education by pointing out the related problems. According to historical records and circumstances such as Sejong's creation of Sinak, Sejo's invention of Jeongganbo with 16 Jeonggan (square) in one vertical line, Sejong's invention of Hangeul and so on, it seems to be the most reasonable that Sejong is the subject of the invention of Jeongganbo as of now. However, the attitude of the musical academy to accept and educate the unclear thing as if it is a fact does not seem desirable. Therefore, I suggest that it should be described "Jeongganbo was invented in the period of Sejong" or "it is supposed that Jeongganbo was invented by Sejong" rather than presenting "Sejong made Jeongganbo" or "created" until revealing the clear evidence about the subject of Jeongganbo.