• Smart Media Journal
• /
• v.7 no.2
• /
• pp.34-39
• /
• 2018

There are many studies on red tide due to the continuous increase in damage to domestic fish and shell farms by the harmful red tide. However, there is insufficient domestic research of identifying harmful red tide algae that automatically recognizes red tide images. In this paper, we propose a red tide image classification method using deep learning based open source. To solve the problem of recognition of various images of red tide algae, the proposed method is implemented by using tensorflow framework and Google image classification model.

• Journal of the Korean Society for Library and Information Science
• /
• v.42 no.4
• /
• pp.181-199
• /
• 2008

This study presents an empirical study of classification system design focused upon an image design team within an organizational setting. It aims to understand collaboration during design practice. Data was collected through on-site interviews, observations, and document and email reviews. This study uses social process model as a conceptual framework. The study revealed type of collaboration, factors influencing collaboration, influences of collaboration on design practice.

• International Journal of Computer Science & Network Security
• /
• v.21 no.7
• /
• pp.324-330
• /
• 2021

Agriculture is one of the essential needs of human life on planet Earth. It is the source of food and earnings for many individuals around the world. The economy of many countries is associated with the agriculture sector. Lots of diseases exist that attack various fruits and crops. Apple Tree Leaves also suffer different types of pathological conditions that affect their production. These pathological conditions include apple scab, cedar apple rust, or multiple diseases, etc. In this paper, an automatic detection framework based on deep learning is investigated for apple leaves disease classification. Different pre-trained models, VGG16, ResNetV2, InceptionV3, and MobileNetV2, are considered for transfer learning. A combination of parameters like learning rate, batch size, and optimizer is analyzed, and the best combination of ResNetV2 with Adam optimizer provided the best classification accuracy of 94%.

• Journal of Korea Multimedia Society
• /
• v.25 no.8
• /
• pp.1203-1211
• /
• 2022

Melanoma is a skin cancer that starts in pigment-producing cells (melanocytes). The death rates of skin cancer like melanoma can be reduced by early detection and diagnosis of diseases. It is common for doctors to spend a lot of time trying to distinguish between skin lesions and healthy cells because of their striking similarities. The detection of melanoma lesions can be made easier for doctors with the help of an automated classification system that uses deep learning. This study presents a new approach for melanoma classification based on an ensemble of deep convolution neural networks and a Log-Gabor filter. First, we create the Log-Gabor representation of the original image. Then, we input the Log-Gabor representation into a new ensemble of deep convolution neural networks. We evaluated the proposed method on the melanoma dataset collected at Yonsei University and Dongsan Clinic. Based on our numerical results, the proposed framework achieves more accuracy than other approaches.

• International Journal of Computer Science & Network Security
• /
• v.23 no.4
• /
• pp.116-122
• /
• 2023

Sentiment categorization technique be commonly isolated interested in threes significant classifications name Machine Learning Procedure (ML), Lexicon Based Method (LB) also finally, the Hybrid Method. In Machine Learning Methods (ML) utilizes phonetic highlights with apply notable ML algorithm. In this paper, in classification and identification be complete base under in optimizations technique called sequential minimal optimization with Random Forest algorithm (SMORF) for expanding the exhibition and proficiency of sentiment classification framework. The three existing classification algorithms are compared with proposed SMORF algorithm. Imitation result within experiential structure is Precisions (P), recalls (R), F-measures (F) and accuracy metric. The proposed sequential minimal optimization with Random Forest (SMORF) provides the great accuracy.

• Current Optics and Photonics
• /
• v.7 no.6
• /
• pp.597-607
• /
• 2023

Volumetric 3D displays generate voxels to enable users to watch three-dimensional virtual objects from various angles, and they have a significant advantage over other types of 3D displays in terms of realism and the absence of vergence-accommodation conflict (VAC). By virtue of these advantages, various volumetric 3D display technologies incorporating novel approaches have been introduced competitively. As a result, the conventional classification criteria for volumetric 3D technology often fall short in categorizing these innovative methods. In this study, we present an improved classification framework capable of accommodating these new technologies. We expect that a new classification may offer some intuition to identify areas of technical deficiency and contribute to improving the technology.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.39 no.3
• /
• pp.18-29
• /
• 2016

For most organizations, a security infrastructure to protect company's core information and their technology is becoming increasingly important. So various approaches to information security have been made but many security accidents are still taking place. In fact, for many Korean companies, information security is perceived as an expense, not an asset. In order to change this perception, it is very important to recognize the need for information security and to find a rational approach for information security. The purpose of this study is to present a framework for information security strategies of companies. The framework classifies companies into eight types so company can receive help in making decisions for the development of information security strategy depending on the type of company it belongs to. To develope measures to classify the types of companies, 12 information security professionals have done brainstorming, and based on previous studies, among the factors that have been demonstrated to be able to influence the information security of the enterprise, three factors have been selected. Delphi method was applied to 29 security experts in order to determine sub items for each factor, and then final items for evaluation was determined by verifying the content validity and reliability of the components through the SPSS analysis. Then, this study identified characteristics of each type of eight companies from a security perspective by utilizing the developed sub items, and summarized what kind of actual security accidents happened in the past.

• Journal of Korean language education
• /
• v.28 no.1
• /
• pp.79-111
• /
• 2017

This study is a trend analysis study that discusses the current status and directions of research methods of KAP research. The existing trend ana lysis studies dealing with research methods have problems in that the classification criteria of the studies used are rough and different from each other, rendering comparison between studies being difficult, and do not comprehensively cover research methods of diversified KAP research. Therefore, this study examined the research methods of KAP research from a critical point of view and suggested a set of classification criteria and an analysis framework that can be used consistently in classification and analysis of future KAP research methods. Based on the theoretical background of second language studies and applied linguistics, this study revised and supplemented Brown (2015)'s research method types and selected 289 journals and theses/dissertations from 2012 to 2016 and classified them into a new analysis framework. The primary and secondary studies, which are the major categories, were 219 and 70, respectively, so it was confirmed that there were much more primary studies. The primary studies then were subdivided into 128 qualitative research studies, 142 survey research studies, and 23 quantitative research studies, pointing to the trend that survey and qualitative research methods were preferred. In the qualitative research approaches, there were 21 action research studies, which were used the most. In addition, such qualitative research approaches as case studies and narrative inquiries which were difficult to find in the past, have gradually increased, confirming that the diversification of research methods is becoming common. However, there were still many studies that did not explicitly put forward research questions and there were many studies that did not report reliability and effect sizes in quantitative research. Of the 23 quantitative studies, only 50% reported reliability, and only three reported effect sizes. In order to enable systematic reviews (meta-analysis) of quantitative research and expect quality improvement of research in future KAP research, reporting of quantitative research should be done more systematically. This study is meaningful in that a systematic and detailed analysis framework was proposed to classify various research methods in the future and that the problems and directions for improvement of the KAP research methods were discussed through the analysis of the research trend of the KAP studies for the last 5 years.

• Journal of Environmental Science International
• /
• v.29 no.12
• /
• pp.1185-1198
• /
• 2020

This study aimed to suggest a framework for biotope classification and evaluation based on habitat values to rationally establish management areas of national parks. The factors and indicators related to the biotope classification and evaluation were established based on integration from those of previous studies. The decision tree evaluation process was applied to the classification and evaluation of the biotope type level. The evaluation of the biotope group level was carried out to determine the weight and the AUEM (Adding Up Estimation Matrix) was applied for the final grades. As a result, the biotope type of Seolaksan National Park was classified into 43 types and Odaesan National Park was classified into 41 types. Bukhansan National Park, which is located in a metropolitan city, was classified into 49 types. In terms of biotope evaluation, grade III had a ratio of 50.6%, the highest in Seolaksan national park. The ratios of grade I and grade II, which have great ecological-value, were respectively 12.0% and 36.5%. Grade II was 48.2%, the highest ratio in Odaesan National Park. The ratios of grades I and II were 10.8% and 37.9%, respectively. Grade III was 54.8%, the highest ratio in Bukhansan national park, and the ratios of grade I and grade II, which have great ecological-value were, respectively, 11.4% and 25.7%. The biotope values of major national parks were evaluated according to the type focusing on the actual vegetation. This framework can be considered for application to the spatial management of other types of protected areas.

• Proceedings of the KOSOMBE Conference
• /
• v.1998 no.11
• /
• pp.255-256
• /
• 1998

In this paper, we propose the multi-axial classification system using parallel coding method that is systemic and flexible properties for representing dental clinical behavior. The methodology and organization of this thesis as follows. First, an analysis of other classification systems. Second, the domain of medical behavior and axises using selected elements was were determined. Third, the new code system is constructed of these common factors in properties of prediction of hierarchy, brevity, simplicity, flexibility and mnemonic usage. Finally, the framework of classification system for dental was made using multi-axial code system. The result of the this study, the eight bases axis of multi-axial code system is composed and can be basic information of research for construction of classification system of all medical domain.