• Journal of the Korean Society of Industry Convergence
• /
• v.24 no.5
• /
• pp.547-553
• /
• 2021

Dynamic object recognition is an important task for autonomous vehicles. Since dynamic objects exhibit a higher collision risk than static objects, our own trajectories should be planned to match the future state of moving elements in the scene. Time information such as optical flow can be used to recognize movement. Existing optical flow calculations are based only on camera sensors and are prone to misunderstanding in low light conditions. In this regard, to improve recognition performance in low-light environments, we applied a normalization filter and a correction function for Gamma Value to the input images. The low light quality improvement algorithm can be applied to confirm the more accurate detection of Object's Bounding Box for the vehicle. It was confirmed that there is an important in object recognition through image prepocessing and deep learning using YOLO.

• Advances in concrete construction
• /
• v.13 no.1
• /
• pp.11-23
• /
• 2022

Conventional concrete needs some improvement in the mechanical properties, which can be obtained by different admixtures. However, making concrete samples costume always time and money. In this paper, different types of hybrid algorithms are applied to develop predictive models for forecasting compressive strength (CS) of concretes containing metakaolin (MK) and fly ash (FA). In this regard, three different algorithms have been used, namely multilayer perceptron (MLP), radial basis function (RBF), and support vector machine (SVR), to predict CS of concretes by considering most influencers input variables. These algorithms integrated with the grey wolf optimization (GWO) algorithm to increase the model's accuracy in predicting (GWMLP, GWRBF, and GWSVR). The proposed MLP models were implemented and evaluated in three different layers, wherein each layer, GWO, fitted the best neuron number of the hidden layer. Correspondingly, the key parameters of the SVR model are identified using the GWO method. Also, the optimization algorithm determines the hidden neurons' number and the spread value to set the RBF structure. The results show that the developed models all provide accurate predictions of the CS of concrete incorporating MK and FA with R² larger than 0.9972 and 0.9976 in the learning and testing stage, respectively. Regarding GWMLP models, the GWMLP1 model outperforms other GWMLP networks. All in all, GWSVR has the worst performance with the lowest indices, while the highest score belongs to GWRBF.

• International journal of advanced smart convergence
• /
• v.13 no.3
• /
• pp.66-79
• /
• 2024

The model developed in this study holds significant importance in predicting carbon emissions in maritime transport. By utilizing ship data and EEDI (Energy Efficiency Design Index) guidelines, the model presents a highly accurate prediction tool, providing a solid foundation for maximizing operational efficiency and effectively managing carbon emissions in ship operations. The model's accuracy was demonstrated by an R² score of 0.95 and a Mean Absolute Percentage Error (MAPE) of 1.4%. Through SHAP (SHapley Additive exPlanations) and Partial Dependence Plots (PDP), it was identified that Speed Over Ground and relative wind speed are the most significant variables, both showing a positive correlation with increased CO2 emissions. Additionally, environmental factors such as exceeding an average draft of 22(m), a Leeway over 5°, and a current angle exceeding 200° were found to increase emissions significantly. Specific ranges of wind and swell wave angles also notably affected emissions. Conversely, lower pitch, roll, and rudder angle were associated with reduced emissions, indicating that stable ship operation enhances efficiency.

• KIPS Transactions on Computer and Communication Systems
• /
• v.3 no.7
• /
• pp.219-230
• /
• 2014

In the cloud computing environment, cloud service users purchase and use the virtualized resources from cloud resource providers on a pay as you go manner. Typically, there are two billing plans for computing resource allocation adopted by large cloud resource providers such as Amazon, Gogrid, and Microsoft, on-demand and reserved plans. Reserved Virtual Machine(VM) instance is provided to users based on the lengthy allocation with the cheaper price than the one of on-demand VM instance which is based on shortly allocation. With the proper mixture allocation of reserved and on-demand VM corresponding to users' requests, cloud service providers are able to reduce the resource allocation cost. To do this, prior researches about VM allocation scheme have been focused on the optimization approach with the users' request prediction techniques. However, it is difficult to predict the expected demands exactly because there are various cloud service users and the their request patterns are heavily fluctuated in reality. Moreover, the previous optimization processing techniques might require unacceptable huge time so it is hard to apply them to the current cloud computing system. In this paper, we propose the cloud brokering system with the adaptive VM allocation schemes called A3R(Adaptive 3 Resource allocation schemes) that do not need any optimization processes and kinds of prediction techniques. By using A3R, the VM instances are allocated to users in response to their service demands adaptively. We demonstrate that our proposed schemes are able to reduce the resource use cost significantly while maintaining the acceptable Quality of Service(QoS) of cloud service users through the evaluation results.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.4
• /
• pp.755-768
• /
• 2017

As Big Data becomes the core of the fourth industrial revolution, big data-based processing and analysis capabilities are expected to influence the company's future competitiveness. Comparative studies of RHadoop and RHIPE that integrate R and Hadoop environment, have not been discussed by many researchers although RHadoop and RHIPE have been discussed separately. In this paper, we constructed big data platforms such as RHadoop and RHIPE applicable to large scale data and implemented the machine learning algorithms such as multiple regression and logistic regression based on MapReduce framework. We conducted a study on performance and scalability with those implementations for various sample sizes of actual data and simulated data. The experiments demonstrated that our RHadoop and RHIPE can scale well and efficiently process large data sets on commodity hardware. We showed RHIPE is faster than RHadoop in almost all the data generally.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.5
• /
• pp.2724-2731
• /
• 2014

As the future national defense plan of government focus on advanced weapon system, military maintenance facility becomes more important. However, military maintenance facility has been managed by director's experience and simple mathematical calculation until now. Thus, the optimization for the management of military maintenance facility is suggested by more scientistic and logical methods in this study. The study follows the procedure below. First, simulation is designed according to the analysis of military maintenance facility. Second, independent variable and dependent variable are defined for optimization. Independent Variable includes the number of maintenance machine, transportation machine, worker in the details of military maintenance facility operation, and dependent variable involves total maintenance time affected by independent variable. Third, warmup analysis is performed to get warmup period, based on the simulation model. Fourth, the optimal combination is computed with evolution strategy, meta-heuristic, to enhance military maintenance management. By the optimal combination, the management of military maintenance facility can gain the biggest effect against the limited cost. In the future, the multipurpose study, to analyze the military maintenance facility covering various weapon system equipments, will be performed.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.6
• /
• pp.935-940
• /
• 2023

Most text data collected through web scraping for artificial intelligence and big data analysis is generally large and unstructured, so a purification process is required for big data analysis. The process becomes structured data that can be analyzed through a heuristic pre-processing refining step and a post-processing machine refining step. Therefore, in this study, in the post-processing machine refining process, the Korean dictionary and the stopword dictionary are used to extract vocabularies for frequency analysis for word cloud analysis. In this process, "user-defined stopwords" are used to efficiently remove stopwords that were not removed. We propose a methodology for applying the "thesaurus" and examine the pros and cons of the proposed refining method through a case analysis using the "user-defined stop word thesaurus" technique proposed to complement the problems of the existing "stop word dictionary" method with R's word cloud technique. We present comparative verification and suggest the effectiveness of practical application of the proposed methodology.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.06a
• /
• pp.426-431
• /
• 1998

We propose a fuzzy rule based method for structure preserving dimensionality reduction. This method selects a small representative sample and applies Sammon's method to project it. The input data points are then augmented by the corresponding projected(output) data points. The augmented data set thus obtained is clustered with the fuzzy c-means(FCM) clustering algorithm. Each cluster is then translated into a fuzzy rule for projection. Our rule based system is computationally very efficient compared to Sammon's method and is quite effective to project new points, i.e., it has good predictability.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.6-9
• /
• 2005

In this study, the ultra precision centerless grinder for ferrule grinding was designed. As the good-qualified ferrule is required a precise and fine grinding, grinding machine for ferrule must have a high accuracy and a sufficient stiffness. The centerless grinder is composed of the high damping concrete bed, grinding wheel spindle unit, regulating wheel spindle unit, feeding table and dressing unit. For a newly developed centerless grinder, hydrostatic system with high precision feeding and high stiffness was proposed. The grinding and regulating wheel spindle units were composed of hydrostatic spindle and feeding table was hydrostatic table. The prototype of hydrostatic table was manufactured and tested.

• Journal of the Korean Society of Safety
• /
• v.8 no.3
• /
• pp.73-77
• /
• 1993

Automation and mechanization of work make people put the machine into operation and watch the state of operations. In the process of those works, they are apt to have accidents caused by their carelessness. To reduce such accidents, we can practise TOUCH and CALL, which Is to Indicate and ascertain the dangerous parts at every process before performing works. The objectives of thls study are to show quantitatively the efficiency of TOUCH & CALL and to examine the effects of S-R compatibility. The results show that reaction time is longer(0.138-0.279sec.) in case of indicating with finge,s and shouting than that of responding only visually. On the other hand, the error rate decreases by 3.3 times-7times. From this, it is considered to verify quantitative estimation on multiple feedback of TOUCH & CALL.