• Korean Journal of Remote Sensing
• /
• v.35 no.6_1
• /
• pp.959-971
• /
• 2019

CNN (Convolutional Neural Network) is one representative deep learning algorithm, which can extract high-level spatial and spectral features, and has been applied for hyperspectral image classification. However, one significant drawback behind the application of CNNs in hyperspectral images is the high dimensionality of the data, which increases the training time and processing complexity. To address this problem, several CNN based hyperspectral image classification studies have exploited PCA (Principal Component Analysis) for dimensionality reduction. One limitation to this is that the spectral information of the original image can be lost through PCA. Although it is clear that the use of PCA affects the accuracy and the CNN training time, the impact of PCA for CNN based hyperspectral image classification has been understudied. The purpose of this study is to analyze the quantitative effect of PCA in CNN for hyperspectral image classification. The hyperspectral images were first transformed through PCA and applied into the CNN model by varying the size of the reduced dimensionality. In addition, 2D-CNN and 3D-CNN frameworks were applied to analyze the sensitivity of the PCA with respect to the convolution kernel in the model. Experimental results were evaluated based on classification accuracy, learning time, variance ratio, and training process. The size of the reduced dimensionality was the most efficient when the explained variance ratio recorded 99.7%~99.8%. Since the 3D kernel had higher classification accuracy in the original-CNN than the PCA-CNN in comparison to the 2D-CNN, the results revealed that the dimensionality reduction was relatively less effective in 3D kernel.

• Journal of Korean Society for Geospatial Information Science
• /
• v.17 no.1
• /
• pp.21-35
• /
• 2009

KOMPSAT-2 is the seventh high-resolution image satellite in the world that provides both 1m-grade panchromatic images of the GSD and 4m-grade multispectral images of the GSD. It's anticipated to be used across many different areas including mapping, territory monitoring and environmental watch. However, due to the complexity and security concern involved with the use of the MSC, the use of KOMPSAT-2 images are limited in terms of geometric images, such as satellite orbits and detailed mapping information. Therefore, this study aims to produce DEM and orthoimage by using the stereo images of KOMPSAT-2, and to explore the applicability of geo-spatial information with KOMPSAT -2. Orientation interpretations were essential for the production of DEM and orthoimage using KOMPSAT-2 images. In the study, they are performed by utilizing both RPC and GCP. In this study, the orientation interpretations are followed by the generation of DEM and orthoimage, and the analysis of their accuracy based on a 1:5,000 digital map. The accuracy analysis of DEM is performed and the results indicate that their altitudes are, in general, higher than those obtained from the digital map. The altitude discrepancies on plains, hills and mountains are calculated as 1.8m, 7.2m, and 11.9m, respectively. In this study, the mean differences between horizontal position between the orthoimage data and the digital map data are found to be ${\pm}3.081m$, which is in the range of ${\pm}3.5m$, within the permitted limit of a 1:5,000 digital map. KOMPSAT-2 images are used to produce DEM and orthoimage in this research. The results suggest that DEM can be adequately used to produce digital maps under 1:5,000 scale.

• Journal of Korean Society for Geospatial Information Science
• /
• v.23 no.3
• /
• pp.23-29
• /
• 2015

Due to the characteristics of landslide disasters including debris flow, the rapid speed to downward and difficulty to respond or evacuate from them, it is imperative to identify their potential hazards and prepare the reduction plans. However, the current landslide hazards generated by a variety of methods has been raised its accuracy because of the complexity of input data and their analyses, and the simplification of the landslide model. The main objective of this study is, therefore, to evaluate the preliminary landslide hazard based on the identification of geomorphological features. Especially, two methodologies based on the statistics of the directional data, Vector dispersion and Planarity analyses, are used to find some relationships between geomorphological characteristics and the landslide hazard. Results show that both methods well discriminate geomorphological features between stable and unstable domains in the landslide areas. Geomorphological features are closely related to the landslide hazard and it is imperative to maximize their characteristics by adapting multiple models rather than individual model only. In conclusions, the mechanism of landslide is not determined solely by a simple cause but the complex natural phenomenon caused by the interactions of the numerous factors and it is of primary importance to require additional researches for the outbreaking mechanism that are based on various methodologies.

• Korean Journal of Health Education and Promotion
• /
• v.18 no.1
• /
• pp.35-47
• /
• 2001

As the complexity of the our environment is further complicated by advancements in industry and increase in vehicle traffic flow, the incidents of injury causing accidents are on the rise. Consequently, there is increasing emphasis on the importance of systematic and continual safety education for injury preventive behaviors. This study investigates safety related problems of elementary school students based on the PRECEDE model, proposed by Green et al.(1980 Green), to comprehensively identify the requirements of school safety education. The identified requirements were used to diagnose the current state of elementary school safety education through the analysis of multidimensional factors. A questionnaire survey was conducted on 594 sixth grade students from randomly selected 4 schools in Seoul to examine their injury preventive behaviors and to determine the educational diagnosis variables that affect it. The duration of the survey was 3 weeks starting from April 12, 1999 to May 8, 1999. A summary of the survey results are presented below; 1. Situations in which accidents have occurred were, in their order of frequency, ‘during play or sports activities within the school grounds’ was most frequent at 59.6％, ‘during play on local streets’ at 49.5％, and ‘traffic accidents’ at 41.6％. 2. Categorization of the injury preventive behavior showed that ‘not playing at high traffic flow locations such as streets and construction sites’ had the higher level of observance, while ‘wearing of helmets and joint protection devices during playing’ was least observed. 3. Considering injury preventive behaviors in relation to educational diagnosis variables indicated, for predisposing factors, lower ‘perception to injury accidents’ (p〈0.001) combined with higher ‘concerns for injury accidents’(p〈0.001), ‘practice of preventive behavior’(p〈0.001), and ‘the level of safety knowledge’(p〈0.001) resulted in significantly higher observance of injury preventive behaviors. For enabling factors, higher ‘perceived level of the school safety education’ (p〈0.001) and ‘availability of safety education resources’(p〈0.01) indicated significantly higher observance of injury preventive behaviors. For the reinforcing factor, frequent exposure to ‘safety education brochure’ (p〈0.01) and ‘audio-visual material for safety education’(p〈0.01) combined with more ‘regional safety education’ (p〈0.01), ‘home safety education’ (p〈0.01), ‘school safety education’(p〈0.001), and, ‘parents’ observance of preventive behaviors' (p〈0.001) showed significantly higher observance of injury preventive behaviors. 4. An analysis of the factors that affect injury preventive behaviors showed that the enabling factor ‘awareness of school safety education’ had the highest correlation with injury preventive behaviors followed by factors, in their order of significance, ‘practice of preventive behavior’, ‘perception to injury accidents’, ‘level of safety knowledge’, ‘parents’ observances of preventive behaviors', and ‘concerns for injury accidents.’

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.13 no.3
• /
• pp.197-205
• /
• 2020

Korean language has the characteristics that the pronunciation of phoneme units such as vowels and consonants are fixed and the pronunciation associated with a notation does not change, so that foreign learners can approach rather easily Korean language. However, when one pronounces words, phrases, or sentences, the pronunciation changes in a manner of a wide variation and complexity at the boundaries of syllables, and the association of notation and pronunciation does not hold any more. Consequently, it is very difficult for foreign learners to study Korean standard pronunciations. Despite these difficulties, it is believed that systematic analysis of pronunciation errors for Korean words is possible according to the advantageous observations that the relationship between Korean notations and pronunciations can be described as a set of firm rules without exceptions unlike other languages including English. In this paper, we propose a visualization framework which shows the differences between standard pronunciations and erratic ones as quantitative measures on the computer screen. Previous researches only show color representation and 3D graphics of speech properties, or an animated view of changing shapes of lips and mouth cavity. Moreover, the features used in the analysis are only point data such as the average of a speech range. In this study, we propose a method which can directly use the time-series data instead of using summary or distorted data. This was realized by using the deep learning-based technique which combines Self-organizing map, variational autoencoder model, and Markov model, and we achieved a superior performance enhancement compared to the method using the point-based data.

• Journal of Intelligence and Information Systems
• /
• v.16 no.1
• /
• pp.93-116
• /
• 2010

Ensemble approach is applied to the detection modeling of illegal cash accommodation (ICA) that is the well-known type of fraudulent usages of credit cards in far east nations and has not been addressed in the academic literatures. The performance of fraud detection model (FDM) suffers from the imbalanced data problem, which can be remedied to some extent using an ensemble of many classifiers. It is generally accepted that ensembles of classifiers produce better accuracy than a single classifier provided there is diversity in the ensemble. Furthermore, recent researches reveal that it may be better to ensemble some selected classifiers instead of all of the classifiers at hand. For the effective detection of ICA, we adopt ensemble size reduction technique that prunes the ensemble of all classifiers using accuracy and diversity measures. The diversity in ensemble manifests itself as disagreement or ambiguity among members. Data imbalance intrinsic to FDM affects our approach for ICA detection in two ways. First, we suggest the training procedure with over-sampling methods to obtain diverse training data sets. Second, we use some variants of accuracy and diversity measures that focus on fraud class. We also dynamically calculate the diversity measure-Forward Addition and Backward Elimination. In our experiments, Neural Networks, Decision Trees and Logit Regressions are the base models as the ensemble members and the performance of homogeneous ensembles are compared with that of heterogeneous ensembles. The experimental results show that the reduced size ensemble is as accurate on average over the data-sets tested as the non-pruned version, which provides benefits in terms of its application efficiency and reduced complexity of the ensemble.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.5
• /
• pp.2641-2654
• /
• 2014

In the smart media environment, magazine industry has been experiencing a transition to ecosystem of value network, which includes high complexity and ambiguity. Using case study method, this article conducts research on digital convergence, the model of magazine ecosystem and adaptation strategy of global magazine companies. Research findings have it that the way of contents production of global magazines has been based on collaborative production system within communities, expert communities, creative users, media contents companies and magazine platform. The system shows different patterns and characteristics depending on magazine-driven platform, Platform-driven platform or user-driven platform. Collaboration system has been confirmed in various cases: Huffington Post and Zinio which collaborate with media contents companies, Amazon magazines and Bookish with magazine companies, Huffington Post and Wired with expert communities, and Flipboard with creative users and communities. Foreign magazine contents diverge into (paper, electronic, app and web magazine) as they start the lively trades of their contents on the magazine platform. In the area of contents uses, readers employ smart media technology effectively such as cloud computing, artificial intelligence and module individualization, making it possible for the virtuous cycle to remain in the relationship within communities, expert communities and creative users.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.9 no.3
• /
• pp.261-270
• /
• 2019

With the advent of the Fourth Industrial Revolution, multi-variety production in the up-to-date manufacturing environment is proceeding more rapidly whereby production planning and management have been getting more complicated. Moreover, the need to improve production efficiency through effective operation sequencing is further heightened. Fundamentally, the effective operation sequencing can reduce the set-up of the equipment, efficiently utilize the equipment, shorten the set-up time, and ultimately contribute to productivity improvement. This study deals with the problem of efficient operation sequencing in a situation where a single machine performs multiple operations. The complexity of the problem is very high when compared to the case where only one operation is performed on one machine, which is covered in most existing studies. In this paper, we propose an integer programming model to minimize the number of setups. This study aims at minimizing the number of mold replacement times in the process of processing a given production order for the wiring harness manufacturing process, which is one of the components of automobile electric field. In addition, brief case studies are presented to verify the validity of the proposed mathematical model.

• Journal of Internet of Things and Convergence
• /
• v.8 no.6
• /
• pp.55-63
• /
• 2022

In recent years, with the rapid development of large and medium-sized urban rail transit in China, the total operating mileage of high-speed railway and the total number of EMUs(Electric Multiple Units) are rising. The system complexity of high-speed EMU is constantly increasing, which puts forward higher requirements for the safety of equipment and the efficiency of maintenance.At present, the maintenance mode of high-speed EMU in China still adopts the post maintenance method based on planned maintenance and fault maintenance, which leads to insufficient or excessive maintenance, reduces the efficiency of equipment fault handling, and increases the maintenance cost. Based on the intelligent operation and maintenance technology of PHM(prognostics and health management). This thesis builds an integrated PHM platform of "vehicle system-communication system-ground system" by integrating multi-source heterogeneous data of different scenarios of high-speed EMU, and combines the equipment fault mechanism with artificial intelligence algorithms to build a fault prediction model for traction motors of high-speed EMU.Reliable fault prediction and accurate maintenance shall be carried out in advance to ensure safe and efficient operation of high-speed EMU.

• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.10
• /
• pp.323-332
• /
• 2022

As various services using AI technology are being developed, much attention is being paid to AI service production. Recently, AI technology is acknowledged as one of ICT services, a lot of research is being conducted for general-purpose AI service production. In this paper, I describe the research results in terms of systems for AI service production, focusing on the distribution and production of machine learning models, which are the final steps of general machine learning development procedures. Three different Ubuntu systems were built, and experiments were conducted on the system, using data from 2017 validation COCO dataset in combination of different AI models (RFCN, SSD-Mobilenet) and different communication methods (gRPC, REST) to request and perform AI services through Tensorflow serving. Through various experiments, it was found that the type of AI model has a greater influence on AI service inference time than AI machine communication method, and in the case of object detection AI service, the number and complexity of objects in the image are more affected than the file size of the image to be detected. In addition, it was confirmed that if the AI service is performed remotely rather than locally, even if it is a machine with good performance, it takes more time to infer the AI service than if it is performed locally. Through the results of this study, it is expected that system design suitable for service goals, AI model development, and efficient AI service production will be possible.