• The Journal of Korea Robotics Society
• /
• v.19 no.1
• /
• pp.79-91
• /
• 2024

Large warehouses are building automation systems to increase efficiency. However, small warehouses, military bases, and local stores are unable to introduce automated logistics systems due to lack of space and budget, and are handling tasks manually, failing to improve efficiency. To solve this problem, this study designed small loading and unloading equipment that can be mounted on transportation vehicles. The equipment can be controlled remotely and is automatically controlled from the point where pallets loaded with cargo are visible using real-time video from an attached camera. Cargo recognition and control command generation for automatic control are achieved through a newly designed deep learning model. This model is designed to be optimized for loading and unloading equipment and mission environments based on the YOLOv3 structure. The trained model recognized 10 types of palettes with different shapes and colors with an average accuracy of 100% and estimated the state with an accuracy of 99.47%. In addition, control commands were created to insert forks into pallets without failure in 14 scenarios assuming actual loading and unloading situations.

• Journal of Intelligence and Information Systems
• /
• v.22 no.2
• /
• pp.127-142
• /
• 2016

Deep learning model is a kind of neural networks that allows multiple hidden layers. There are various deep learning architectures such as convolutional neural networks, deep belief networks and recurrent neural networks. Those have been applied to fields like computer vision, automatic speech recognition, natural language processing, audio recognition and bioinformatics where they have been shown to produce state-of-the-art results on various tasks. Among those architectures, convolutional neural networks and recurrent neural networks are classified as the supervised learning model. And in recent years, those supervised learning models have gained more popularity than unsupervised learning models such as deep belief networks, because supervised learning models have shown fashionable applications in such fields mentioned above. Deep learning models can be trained with backpropagation algorithm. Backpropagation is an abbreviation for "backward propagation of errors" and a common method of training artificial neural networks used in conjunction with an optimization method such as gradient descent. The method calculates the gradient of an error function with respect to all the weights in the network. The gradient is fed to the optimization method which in turn uses it to update the weights, in an attempt to minimize the error function. Convolutional neural networks use a special architecture which is particularly well-adapted to classify images. Using this architecture makes convolutional networks fast to train. This, in turn, helps us train deep, muti-layer networks, which are very good at classifying images. These days, deep convolutional networks are used in most neural networks for image recognition. Convolutional neural networks use three basic ideas: local receptive fields, shared weights, and pooling. By local receptive fields, we mean that each neuron in the first(or any) hidden layer will be connected to a small region of the input(or previous layer's) neurons. Shared weights mean that we're going to use the same weights and bias for each of the local receptive field. This means that all the neurons in the hidden layer detect exactly the same feature, just at different locations in the input image. In addition to the convolutional layers just described, convolutional neural networks also contain pooling layers. Pooling layers are usually used immediately after convolutional layers. What the pooling layers do is to simplify the information in the output from the convolutional layer. Recent convolutional network architectures have 10 to 20 hidden layers and billions of connections between units. Training deep learning networks has taken weeks several years ago, but thanks to progress in GPU and algorithm enhancement, training time has reduced to several hours. Neural networks with time-varying behavior are known as recurrent neural networks or RNNs. A recurrent neural network is a class of artificial neural network where connections between units form a directed cycle. This creates an internal state of the network which allows it to exhibit dynamic temporal behavior. Unlike feedforward neural networks, RNNs can use their internal memory to process arbitrary sequences of inputs. Early RNN models turned out to be very difficult to train, harder even than deep feedforward networks. The reason is the unstable gradient problem such as vanishing gradient and exploding gradient. The gradient can get smaller and smaller as it is propagated back through layers. This makes learning in early layers extremely slow. The problem actually gets worse in RNNs, since gradients aren't just propagated backward through layers, they're propagated backward through time. If the network runs for a long time, that can make the gradient extremely unstable and hard to learn from. It has been possible to incorporate an idea known as long short-term memory units (LSTMs) into RNNs. LSTMs make it much easier to get good results when training RNNs, and many recent papers make use of LSTMs or related ideas.

• The Journal of Art Theory & Practice
• /
• no.9
• /
• pp.33-57
• /
• 2010

Focused on the performance of interactivity in the appreciation of media art, this study is associated with the continually changed art texts and a participator's physical reactions to savor the art. Digital works of art emphasizing interactivity usually decentralize and disperse the responsibility and the central role for creating and producing the art works. Proper procedure of the work is generated by actions such as clicking the mouse, controlling the joy stick or actual movement of anticipators' bodies. The art works are influenced by participators' interactivity, which makes the leading roles and the responsibility for creating art scattered and divided. These features are similar with those of the 'Rhizome' which Gilles Deleuze(1925~1995) and Felix Guattari(1930~1992) have discussed. In their argument, 'Rhizome' is an interval or being 'between', which keeps changing. 'Rhizome' is a state in which the individual and the work of art never reach the conclusion, only a phenomenon of eternally altering. Like 'Rhizome', this sort of art work has the decentralized system, opens for several directions, and activates the system which is changeable as linked items increased or decreased. These works stimulate the individual to perform and act while appreciating the art piece. In terms of processing and preceeding, interactivity is the important equipment and catalyzer. Through these procedures, the pieces can be the 'floating work of art' combined and condensed with the whole participators' reactions. The 'floating work of art' is neither the expression of an individual nor that of one particular group. Multidimensional influence of the web is the web which is constantly reorganizing and producing in its connective state. This connective state is activated by interactivity. The Rhizomatic system embodies the floating work of art process. Due to each individual perceiving art in individualistic terms, there is no dominating powers or central points. I regard this art works possessing above traits as the work of art with Rhizomatic system. The work of art with the Rhizomatic system is embodied through interactivity and because physical action activates the process of appreciation, a participator can actually experience and practice the philosophy. Ultimately the Rhizomatic speculation occurs during the interactivity of appreciating the Rhizomatic art pieces. The Rhizomatic system penetrates into the intuitive area beyond our recognition and thoughts, as we are engaged in the connective process. With the methods and manners of interactive art, we can possibly reconsider the system as a tool in which the participator is directly able to link experience and theory to the philosophies of Deleuze and Guattari.

• The Journal of the Korea Contents Association
• /
• v.18 no.4
• /
• pp.550-566
• /
• 2018

In the era of the 4th industrial revolution, enhancing achievement evaluation based on process oriented instruction is essential. It assists human growth both cognitively and mentally. The purpose of this study is to analyze current condition of high school achievement evaluation in Daejeon region for enhancing English class based on achievement Assessment. Analyzing high school achievement evaluation plans as well as analyzing items of paper and pencil test using TELL program was conducted. Reanalysis of global citizen theme-based English according to core achievement standard was performed. The questionnaire was analyzed using the SPSS Win 20.0 Program to figure out significant difference of instructional method, the rate of students' grade improvement, English class related to the state of career recognition. T-test, ANOVA was performed to determine if there was a difference between the individual instructional variables. The research result is designed to construct or develop English class based on achievement evaluation while providing each high school with the result of current state of high school achievement evaluation. Specific characteristics of individual achievement result was conducted in terms of analyzing distribution of answer sheet response in order to be used as information for managing each high school achievement evaluation.

• The Journal of the Korea Contents Association
• /
• v.17 no.12
• /
• pp.545-557
• /
• 2017

In this study, as Korea's new government began, the voice of the so-called "real country" has increased, and what has been said about the "good country" and "good government". Therefore, based on the statements and researches on 'good country' and 'good government' that we talk about in our daily life, we collected associative statements about 'good country'. As a result of analysis, we classified the 'good countries' into 5 types as follows. is named "Trusted State Type" and is named "Workable State Type". is called "national type for children," is named "happy national type," and is named "living type." According to the results of the analysis, there is no significant difference in recognition of good countries according to age and occupation. In the twenty-first century, the Republic of Korea is forced to rethink its understanding of our "state" and the reason for its existence as a new government is introduced in a period of turbulence called regime change. This study intends to give meaning to the meaning of existence of 'Government' and opportunity to recall the desire and expectation of 'Good Government'

• Agricultural and Biosystems Engineering
• /
• v.4 no.1
• /
• pp.8-15
• /
• 2003

In Korea, quality evaluation of dried oak mushrooms are done first by classifying them into more than 10 different categories based on the state of opening of the cap, surface pattern, and colors. And mushrooms of each category are further classified into 3 or 4 groups based on its shape and size, resulting into total 30 to 40 different grades. Quality evaluation and sorting based on the external visual features are usually done manually. Since visual features of mushroom affecting quality grades are distributed over the entire surface of the mushroom, both front (cap) and back (stem and gill) surfaces should be inspected thoroughly. In fact, it is almost impossible for human to inspect every mushroom, especially when they are fed continuously via conveyor. In this paper, considering real time on-line system implementation, image processing algorithms utilizing artificial neural network have been developed for the quality grading of a mushroom. The neural network based image processing utilized the raw gray value image of fed mushrooms captured by the camera without any complex image processing such as feature enhancement and extraction to identify the feeding state and to grade the quality of a mushroom. Developed algorithms were implemented to the prototype on-line grading and sorting system. The prototype was developed to simplify the system requirement and the overall mechanism. The system was composed of automatic devices for mushroom feeding and handling, a set of computer vision system with lighting chamber, one chip microprocessor based controller, and pneumatic actuators. The proposed grading scheme was tested using the prototype. Network training for the feeding state recognition and grading was done using static images. 200 samples (20 grade levels and 10 per each grade) were used for training. 300 samples (20 grade levels and 15 per each grade) were used to validate the trained network. By changing orientation of each sample, 600 data sets were made for the test and the trained network showed around 91 % of the grading accuracy. Though image processing itself required approximately less than 0.3 second depending on a mushroom, because of the actuating device and control response, average 0.6 to 0.7 second was required for grading and sorting of a mushroom resulting into the processing capability of 5,000/hr to 6,000/hr.

• Journal of Service Research and Studies
• /
• v.6 no.3
• /
• pp.1-15
• /
• 2016

Despite that Korea has tried radical efforts in the global flow of regulatory reform in the past twenty years, the result was not radical at all, but rather disappointing. One examines the possibility of paradigm shift in regulatory reform based on new theoretic perspectives. Regulatory reform, one argues, is not just a neo-liberal approach to cut off overflowing regulation. It is a highly conflictual struggle in state order to move from industrial age paradigm to service age paradigm. In the process of the great shift states become integrated into the world of life constructed by the exercise of civil rights. The relation between the civic socio-economic life and the state apparatus became totally different. Past effort for deregulation missed this point without correct recognition of the role of civil freedom and rights in service economy. One treats three typical forms of regulation whereby conventional rules and regulations effectively damper the development of services: reciprocal perspective in contract management, industrial mind in urban and spaces design, and old way of human capital management. According these analyses a new initiative of regulatory reform is proposed to take place at the National Assembly.

• Journal of Korean Public Health Nursing
• /
• v.12 no.1
• /
• pp.132-146
• /
• 1998

This Study was attempted to provide the basic data necessary in the development and introduction of Home Health Nursing System by investigating the recognition and attitude level of Home Health Nursing System. The data were collected by means of questionaires presented to 74 patients who had been admitted in C general hospital in Chon Ju, from June 30, 1997. As the tool for this study, the questionares developed by Kim Yong. Soon, et al (1990) and Han Bok Hee(1993) were modified and supplemented for the aim of this study. The computer was used for data analysis. The items about the charateristics of the subjects and the attitude to the management plan of Home Health Nursing System were represented as the frequency and percentage. The standard deviation and calculation average were produced on the items related to definition, recognition, necessity, expected effect of the attitude of Home Health Nursing System and the items related to admission. The ANOVA test was .used according to the characteristics of variables to analyze the necessity and difference of Home Health Nursing System. The results of this study were as follows 1) The general characteristics of the subjects were as follows ; for sex, man, $58.1\%$ ; for age, 50-59 years, $29.7\%$ ; for the level of education, high school, $51.4\%$ ; $79.7\%$ of them were married; for the family forms, small family, $73.0\%$ ; and $68.9\%$ of them take the monthly income over 100 million won. 2) The characteristics related to admissions of the subjects were as follows ; for clinic, surgical department, $78.4\%$ ; addmission not more then 7days, $47.3\%$ ; for the operation-performance $71.6\%$ of them were experienced; for the admission route, via outpatients clinic, $54.1\%$ ; for waiting period to the admission day, 1-2 days, $71.6\%$. 3) The difficulties comming from the hospitalization were related mostly to the factor that they felt hospital life more inconvenient than home.(3.66) The reasons for the difficulties in the admission which was due to insufficient beds in the hospital was related to the concentration to the general hospital because of 'The Whole National Medical Insurance System'(4.05). 4) On the previous informations about the Home Health Nursing System, those who have heard of only the name were 42 $(56.8\%)$, and on the recognition of it, they thought that it is periodic treatment by the licenced nurses for the recovering pateints after early discharge(3.73). On the attitude about the necessity of Home Health Nursing System, they thought that it is necessary because of the increasing trend of a psychological disease by the change of environment and complexity of the social structure(4.24). On the expected effect of Home Health Nursing System, they answered that it is convinient for the family of the patient to take care of them(4.l8). 5) On the attitude to the management plan of the Home Health Nursing System, those who had intention to participate in the system in the case of systemic support were 42(56.8). In the visiting time, 'visit periodically' and 'visit when the patient needs' were $28(37.8\%)$ respectively. For the application of medical insurance, if possoble, they will use $(91.9\%)$; for the method of payment for the treatment, 'pay by the time required' was $23(31.1\%)$, for the subject of management, 'National public institute must operate' was $33(44.6\%)$. 6) The relationship between the general characteristics of the subjects and the necessity of Home Health Nursing System showed the notable difference in the age (F=3.508, P<0.05) and marrage state (F=5.402, P<.023).

• Journal of Arbitration Studies
• /
• v.17 no.3
• /
• pp.3-30
• /
• 2007

The New York Convention(formally called "United Nations Convention on the Recognition and Enforcement of Foreign Arbitral Awards") done in New York on June 10, 1958 has been adhered to by more than 140 States at the time of this writing, including almost all important trading nations from the Capitalist and Socialist World as well as many developing countries. The Convention can be considered as the most important Convention in the field of arbitration and as the cornerstone of current international commercial arbitration. Korea has acceded to the New York Convention since 1973. When acceding to the Convention, Korea declared that it will apply the Convention to the recognition and enforcement of awards made only in the territory of another Contracting State on the basis of reciprocity. Also, Korea declared that it will apply the Convention only to differences arising out of legal relationships, whether contractual or not, which are considered as commercial under the national law of Korea. The provisions relating to the enforcement of arbitral awards falling under the New York Convention begin at Article III. The Article III contains the general obligation for the Contracting States to recognize Convention awards as binding and to enforce them in accordance with their rules of procedure. The Convention requires a minimum of conditions to be fulfilled by the party seeking enforcement. According to Article IV(1), that party has only to supply (1) the duly authenticated original award or a duly certified copy thereof, and (2) the original arbitration agreement or a duly certified copy thereof. In fulfilling these conditions, the party seeking enforcement produces prima facie evidence entitling it to obtain enforcement of the award. It is then up to the other party to prove that enforcement should not be granted on the basis of the grounds for refusal of enforcement enumerated in the subsequent Article V(1). Grounds for refusal of enforcement are stipulated in Article V is divided into two parts. Firstly, listed in the first Para. of Article V are the grounds for refusal of enforcement which are to be asserted and proven by the respondent. Secondly, listed in Para. 2 of Article V, are the grounds on which a court may refuse enforcement on its own motion. These grounds are non-arbitrability of the subject matter and violation of the public policy of the enforcement country. The three main features of the grounds for refusal of enforcement of an award under Article V, which are almost unanimously affirmed by the courts, are the following. Firstly, The grounds for refusal of enforcement mentioned in Article V are exhaustive. No other grounds can be invoked. Secondly, and this feature follows from the first one, the court before which enforcement of the award is sought may not review the merits of the award because a mistake in fact or law by the arbitrators is not included in the list of grounds for refusal of enforcement set forth in Article V. Thirdly, the party against whom enforcement is sought has the burden of proving the existence of one or more of the grounds for refusal of enforcement. The grounds for refusal of enforcement by a court on its own motion, listed in the second Para. of Article V, are non-arbitrability of the subject matter and public policy of the enforcement country. From the court decisions reported so far at home and abroad, it appears that courts accept a violation of public policy in extreme cases only, and frequently justify their decision by distinguishing between domestic and international public policy. The Dec. 31, 1999 amendment to the Arbitration Act of Korea admits the basis for enforcement of foreign arbitral awards rendered under the New York Convention. In Korea, a holder of a foreign arbitral award is obliged to request from the court a judgment ordering enforcement of the award.

• Journal of the Korean School Mathematics Society
• /
• v.11 no.4
• /
• pp.609-627
• /
• 2008

The purpose of this research was to investigate the relationship between the students' strategy types and the recognition for proportional situations. The students' strategy types which were based on the results of ratio and proportion tests were divided into an additive type, a multiplicative type, and a formal type. This research analyzed the students' activities of categorization when were given the proportional problems and nonproportional problems to the students. And it also explored how to develop students' recognizing for the discrimination between the proportional situations and nonproportional situations. The results was the following. First, the students didn't discriminate the proportional situations and the nonproportional situations in the initial state but they came to discriminate little by little. Secondly, the students didn't discriminate the direct proportions and the inverse proportions until the last stage. Third, the multiplicative type was outperformed more than the formal type in solving the ratio and proportion problems but the formal type was outperformed more than the multiplicative type in discriminating between proportional situations and nonproportional situations. These results are interpreted as showing that solving ratio and proportion tasks and recognizing proportional situations are different aspects of proportional reasoning and it is necessary to understand multiplicative strategy with formal strategy in recognizing proportional situations.