• Korean Journal of Remote Sensing
• /
• v.35 no.1
• /
• pp.93-115
• /
• 2019

The availability of high-resolution laser scanning data and advanced machine learning algorithms has enabled an accurate potential rockfall source identification. However, the presence of other mass movements, such as landslides within the same region of interest, poses additional challenges to this task. Thus, this research presents a method based on an integration of Gaussian mixture model (GMM) and ensemble artificial neural network (bagging ANN [BANN]) for automatic detection of potential rockfall sources at Kinta Valley area, Malaysia. The GMM was utilised to determine slope angle thresholds of various geomorphological units. Different algorithms(ANN, support vector machine [SVM] and k nearest neighbour [kNN]) were individually tested with various ensemble models (bagging, voting and boosting). Grid search method was adopted to optimise the hyperparameters of the investigated base models. The proposed model achieves excellent results with success and prediction accuracies at 95% and 94%, respectively. In addition, this technique has achieved excellent accuracies (ROC = 95%) over other methods used. Moreover, the proposed model has achieved the optimal prediction accuracies (92%) on the basis of testing data, thereby indicating that the model can be generalised and replicated in different regions, and the proposed method can be applied to various landslide studies.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.5
• /
• pp.1778-1797
• /
• 2021

Document image binarization is an important pre-processing step in document analysis and archiving. The state-of-the-art models for document image binarization are variants of encoder-decoder architectures, such as FCN (fully convolutional network) and U-Net. Despite their success, they still suffer from three limitations: (1) reduced feature map resolution due to consecutive strided pooling or convolutions, (2) multiple scales of target objects, and (3) reduced localization accuracy due to the built-in invariance of deep convolutional neural networks (DCNNs). To overcome these three challenges, we propose an improved semantic segmentation model, referred to as DP-LinkNet, which adopts the D-LinkNet architecture as its backbone, with the proposed hybrid dilated convolution (HDC) and spatial pyramid pooling (SPP) modules between the encoder and the decoder. Extensive experiments are conducted on recent document image binarization competition (DIBCO) and handwritten document image binarization competition (H-DIBCO) benchmark datasets. Results show that our proposed DP-LinkNet outperforms other state-of-the-art techniques by a large margin. Our implementation and the pre-trained models are available at https://github.com/beargolden/DP-LinkNet.

• Wind and Structures
• /
• v.34 no.6
• /
• pp.499-510
• /
• 2022

Computational Wind Engineering has rapidly grown in the last decades and it is currently reaching a relatively mature state. The prediction of wind loading by means of numerical simulations has been proved effective in many research studies and applications to design practice are rapidly spreading. Despite such success, caution in the use of simulations for wind loading assessment is still advisable and, indeed, required. The computational burden and the know-how needed to run high-fidelity simulations is often unavailable and the possibility to use simplified models extremely attractive. In this paper, the applicability of some well-known 2D unsteady RANS models, particularly the k-ω SST, in the aerodynamic characterization of extruded bodies with bluff sections is investigated. The main focus of this paper is on the drag coefficient prediction. The topic is not new, but, in the authors' opinion, worth a careful revisitation. In fact, despite their great technical relevance, a systematic study focussing on sections which manifest a fully detached flow configuration has been overlooked. It is here shown that the considered 2D RANS exhibit a pathological behaviour, failing to reproduce the transition between reattached and fully detached flow regime.

• The Journal of Korea Robotics Society
• /
• v.16 no.4
• /
• pp.299-305
• /
• 2021

In this study, the task of robotic tidy-up is to clean the current environment up exactly like a target image. To perform a tidy-up task using a robot, it is necessary to estimate the pose of various objects and to classify the objects. Pose estimation requires the CAD model of an object, but these models of most objects in daily life are not available. Therefore, this study proposes an algorithm that uses point cloud and PCA to estimate the pose of objects without the help of CAD models in cluttered environments. In addition, objects are usually detected using a deep learning-based object detection. However, this method has a limitation in that only the learned objects can be recognized, and it may take a long time to learn. This study proposes an image matching based on few-shot learning and Siamese network. It was shown from experiments that the proposed method can be effectively applied to the robotic tidy-up system, which showed a success rate of 85% in the tidy-up task.

• International Journal of Computer Science & Network Security
• /
• v.23 no.5
• /
• pp.73-88
• /
• 2023

Brain tumors can also be an abnormal collection or accumulation of cells in the brain that can be life-threatening due to their ability to invade and metastasize to nearby tissues. Accurate diagnosis is critical to the success of treatment planning, and resonant imaging is the primary diagnostic imaging method used to diagnose brain tumors and their extent. Deep learning methods for computer vision applications have shown significant improvements in recent years, primarily due to the undeniable fact that there is a large amount of data on the market to teach models. Therefore, improvements within the model architecture perform better approximations in the monitored configuration. Tumor classification using these deep learning techniques has made great strides by providing reliable, annotated open data sets. Reduce computational effort and learn specific spatial and temporal relationships. This white paper describes transfer models such as the MobileNet model, VGG19 model, InceptionResNetV2 model, Inception model, and DenseNet201 model. The model uses three different optimizers, Adam, SGD, and RMSprop. Finally, the pre-trained MobileNet with RMSprop optimizer is the best model in this paper, with 0.995 accuracies, 0.99 sensitivity, and 1.00 specificity, while at the same time having the lowest computational cost.

• International Journal of Advanced Culture Technology
• /
• v.11 no.2
• /
• pp.10-29
• /
• 2023

Grit or perseverance as a factor for student success and life has gained increasing attention. Statistical methods have been the norm in analyzing various aspects of grit, but they do not address the transient and dynamic behavior well. We, for the first time, developed two linear dynamical models that specifically address the feedback structure of a child's desire to achieve a high grade point average (GPA) and the necessary effort that will increase stress between parents and a child. We call the dynamical model as GSES (Goal, Status, Effort & Stress). The two dynamical models incorporate the positive (i.e., achieving a high GPA) and the negative sides (i.e., effort and elevated stress and thus unhappiness) for being gritty or perseverant. Different types of parenting style and a child's characteristics were simulated whether parents and a child are empathetic or stubborn to their expectations and stress (i.e., willing or unwilling to change). Simulations show that when both parents and a child are empathetic to each other's expectation and stress, the most stable situations with minimal stress and effort occur. When a stubborn parent's and a stubborn child were studied together, this resulted in the highest elevation of stress and effort. Stubborn parents and a complying or empathetic child resulted in considerably high stress to a child. Interference from parents may unexpectedly result in a situation in which a child's stress is seriously elevated. The GSES model shows the U-shaped happiness curve (i.e., reciprocal of stress) caused by the increasing and then decreasing goal

• Communications for Statistical Applications and Methods
• /
• v.30 no.6
• /
• pp.589-603
• /
• 2023

Deep generative models target to infer the underlying true data distribution, and it leads to a huge success in generating fake-but-realistic data. Regarding such a perspective, the data attributes can be a crucial factor in the data generation process since non-existent counterfactual samples can be generated by altering certain factors. For example, we can generate new portrait images by flipping the gender attribute or altering the hair color attributes. This paper proposes counterfactual disentangled variational autoencoder generative adversarial networks (CDVAE-GAN), specialized for data attribute level counterfactual data generation. The structure of the proposed CDVAE-GAN consists of variational autoencoders and generative adversarial networks. Specifically, we adopt a Gaussian variational autoencoder to extract low-dimensional disentangled data features and auxiliary Bernoulli latent variables to model the data attributes separately. Also, we utilize a generative adversarial network to generate data with high fidelity. By enjoying the benefits of the variational autoencoder with the additional Bernoulli latent variables and the generative adversarial network, the proposed CDVAE-GAN can control the data attributes, and it enables producing counterfactual data. Our experimental result on the CelebA dataset qualitatively shows that the generated samples from CDVAE-GAN are realistic. Also, the quantitative results support that the proposed model can produce data that can deceive other machine learning classifiers with the altered data attributes.

• Earthquakes and Structures
• /
• v.27 no.2
• /
• pp.83-95
• /
• 2024

The prediction of seismic behavior of the existing building stock is one of the most impactful and complex problems faced by countries with frequent and intense seismic activities. Human lives can be threatened or lost, the economic life is disrupted and large amounts of monetary reparations can be potentially required. However, authorities at a regional or national level have limited resources at their disposal in order to allocate to preventative measures. Thus, in order to do so, it is essential for them to be able to rank a given population of structures according to their expected degree of damage in an earthquake. In this paper, the authors present a ranking approach, based on Machine Learning (ML) algorithms for pairwise comparisons, coupled with ad hoc ranking rules. The case study employed data from 404 reinforced concrete structures with various degrees of damage from the Athens 1999 earthquake. The two main components of our experiments pertain to the performance of the ML models and the success of the overall ranking process. The former was evaluated using the well-known respective metrics of Precision, Recall, F1-score, Accuracy and Area Under Curve (AUC). The performance of the overall ranking was evaluated using Kendall's tau distance and by viewing the problem as a classification into bins. The obtained results were promising, and were shown to outperform currently employed engineering practices. This demonstrated the capabilities and potential of these models in identifying the most vulnerable structures and, thus, mitigating the effects of earthquakes on society.

• Journal of Distribution Science
• /
• v.13 no.2
• /
• pp.55-62
• /
• 2015

Purpose - This study aims to investigate the success factors pertaining to K-POPs from an analysis of the internal business environment. Much research has investigated Korean Moves or how to popularize them. The research mainly focused on aspects of Korean Moves. However, few studies have attempted to examine Korean Moves or K-POPs from a managerial viewpoint. The current research tries to investigate the success factors of K-POP from strategic perspectives, specifically utilizing internal resource based view perspectives. It differentiates itself by looking at the competitiveness of K-POP from the internal resources. Research design, data, and methodology - In the entertainment industry, where creativity is heavily stressed, competitiveness is often regarded within the organization as a form of intangible asset, knowledge, or technology that is often related with the organization's personnel. Some research has tried to reveal the competitiveness of K-POP using Porter's competitiveness of nations framework. Others utilize the adapted model of Porter's structure. However, these models only look at the outside environment, and not inside a firm's resource, knowledge, or capabilities. This research utilizes the VRIO model to examine the internal resources and capabilities of K-POP producers. The model measures whether a firm's internal resources and capabilities are valuable, rare, difficult to imitate by competitors, or organizable. The research covered businesses whose yearly revenue exceeds $10 Million in music planning and recording in South Korea. There were only thirteen such companies (one percent of the total population). Of these, companies for whom 20 percent or more of the sales revenue comes from the abroad are targeted. Only seven are selected and these participated in the research. In order to find a firm's internal resources, we conducted qualitative research methodology. Their business names and persons who participated in this research are not revealed due to case sensitive issues. Instead, we use unrelated initials for their names and their statements. Results - From the in-depth interview with top-tier K-POP producers and managers, the current research tried to identify resources and capabilities that helped to strengthen their competitiveness. These resources and capabilities are sought from the scope of the VRIO model, which looks at the internal resources and capabilities from the scope of value, rarity, imitability, and organization. Interviews with the top tier producers and managers reveal the internal success factors of K-POPs. We conclude that these resources and capabilities are from internally accumulated producing know-how, unique managing (training) system, and outstanding all-round entertainment capabilities of the performers. Conclusions - These results indicate that the core resources and capabilities of K-POP are robust. It will take a significant amount of time and money to imitate for followers, because these resources and capabilities are the result of time investment and are embedded into producers' and performers' know-how. Taking Luo (2000)'s argument, K-POP is in the second stage of the globalization process, which is configuring and allocation resource capabilities to a global scope.

• East Asian Journal of Business Economics (EAJBE)
• /
• v.5 no.4
• /
• pp.39-49
• /
• 2017

Purpose - This study is to propose a creative idea for constant business growth and development by examining characteristics of business outcomes by phase, which are "growth" and "erosion and stagnation," respectively. Research design, data, methodology - It is necessary to identify an occurrence of crisis and its diffusion with a dynamic model in order to identify a success and failure of businesses in an organic way, not on a binary structure. The static perspective is to understand a crisis as a simply one-time event or as a linear causation. Thus, it has a limited understanding of the overall situation and has limits to investigating a foundational cause and developing long-term countermeasures. On the contrary, the dynamic perspective is to understand the crisis as circulation process of the overall system. Thus, it divides elements of the crisis as external and internal ones to understand it as the causal relationship of each element. Results - During the growth period of Burger King, the company promoted its brand very successfully with aggressive and creative marketing activities. However, due to the founder's disposal of management rights and the following changes in the management, the company had no choice but to lose focus on its business philosophy and brand management, and eventually it had to face the big crisis (resonance) which was delisting from the stock market because of the external threat; well-being trend. However, Burger King resumed lifting on the stock exchange by making great efforts to clearly identify the current issues and seek solutions. Under the spirit of "perseverance" and its slogan "Have it your way" the company is now going head to head with McDonald's in the North American region and emerging countries. Conclusions - Then, what is the most crucial factor in the success and failure of businesses? Answers may vary, however, as learned from the case study of Burger King, corporations should inspect the present and focus on developing a long-term strategy for the future and actively fulfill the actions. McDonald's may not be able to innovate by itself in the future as it may become routinized to the growth. There will be chances of winning if we change conditions of individuals or organizations to an organic system in terms of being creative. There is a hopeful message here that an individual or small business may have more advantages in the era of the idea and innovation.