• Journal of KIISE:Computer Systems and Theory
• /
• v.31 no.5_6
• /
• pp.297-304
• /
• 2004

In this paper, a mesh simplification algorithm based on wavelet transform and 2D planar sampling is proposed for efficient handling of 3D objects in computer applications. Since 3D vertices are directly transformed with wavelets in conventional mesh compression and simplification algorithms, it is difficult to solve tiling optimization problems which reconnect vertices into faces in the synthesis stage highly demanding vertex connectivities. However, a 3D mesh is sampled onto 2D planes and 2D polygons on the planes are independently simplified in the proposed algorithm. Accordingly, the transform of 2D polygons is very tractable and their connection information Is replaced with a sequence of vertices. The vertex sequence of the 2D polygons on each plane is analyzed with wavelets and the transformed data are simplified by removing small wavelet coefficients which are not dominant in the subjective quality of its shape. Therefore, the proposed algorithm is able to change the mesh level-of-detail simply by controlling the distance of 2D sampling planes and the selective removal of wavelet coefficients. Experimental results show that the proposed algorithm is a simple and efficient simplification technique with less external distortion.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.5
• /
• pp.7-14
• /
• 2021

Tomato crops are highly affected by tomato diseases, and if not prevented, a disease can cause severe losses for the agricultural economy. Therefore, there is a need for a system that quickly and accurately diagnoses various tomato diseases. In this paper, we propose a system that classifies nine diseases as well as healthy tomato plants by applying various pretrained deep learning-based CNN models trained on an ImageNet dataset. The tomato leaf image dataset obtained from PlantVillage is provided as input to ResNet, Xception, and DenseNet, which have deep learning-based CNN architectures. The proposed models were constructed by adding a top-level classifier to the basic CNN model, and they were trained by applying a 5-fold cross-validation strategy. All three of the proposed models were trained in two stages: transfer learning (which freezes the layers of the basic CNN model and then trains only the top-level classifiers), and fine-tuned learning (which sets the learning rate to a very small number and trains after unfreezing basic CNN layers). SGD, RMSprop, and Adam were applied as optimization algorithms. The experimental results show that the DenseNet CNN model to which the RMSprop algorithm was applied output the best results, with 98.63% accuracy.

• Journal of Broadcast Engineering
• /
• v.26 no.3
• /
• pp.247-257
• /
• 2021

This paper proposes a new technique for calibrating a multi-view RGB-D camera using a 3D (dimensional) skeleton. In order to calibrate a multi-view camera, consistent feature points are required. In addition, it is necessary to acquire accurate feature points in order to obtain a high-accuracy calibration result. We use the human skeleton as a feature point to calibrate a multi-view camera. The human skeleton can be easily obtained using state-of-the-art pose estimation algorithms. We propose an RGB-D-based calibration algorithm that uses the joint coordinates of the 3D skeleton obtained through the posture estimation algorithm as a feature point. Since the human body information captured by the multi-view camera may be incomplete, the skeleton predicted based on the image information acquired through it may be incomplete. After efficiently integrating a large number of incomplete skeletons into one skeleton, multi-view cameras can be calibrated by using the integrated skeleton to obtain a camera transformation matrix. In order to increase the accuracy of the calibration, multiple skeletons are used for optimization through temporal iterations. We demonstrate through experiments that a multi-view camera can be calibrated using a large number of incomplete skeletons.

• Journal of Navigation and Port Research
• /
• v.46 no.3
• /
• pp.280-288
• /
• 2022

This study examined the diagnostics of abnormalities and faults of equipment, whose rotational speed changes even during regular operation. The purpose of this study was to suggest a procedure that can properly apply machine learning to the time series data, comprising non-stationary characteristics as the rotational speed changes. Anomaly and fault diagnosis was performed using machine learning: k-Nearest Neighbor (k-NN), Support Vector Machine (SVM), and Random Forest. To compare the diagnostic accuracy, an autoencoder was used for anomaly detection and a convolution based Conv1D was additionally used for fault diagnosis. Feature vectors comprising statistical and frequency attributes were extracted, and normalization & dimensional reduction were applied to the extracted feature vectors. Changes in the diagnostic accuracy of machine learning according to feature selection, normalization, and dimensional reduction are explained. The hyperparameter optimization process and the layered structure are also described for each algorithm. Finally, results show that machine learning can accurately diagnose the failure of a variable-rotation machine under the appropriate feature treatment, although the convolution algorithms have been widely applied to the considered problem.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.26 no.6
• /
• pp.157-166
• /
• 2022

Structural supervision on the construction site has been performed based on visual inspection, which is highly labor-intensive and subjective. In this study, the remote technique was developed to improve the efficiency of the measurements on rebar spacing using a 360° camera and reconstructed 3D models. The proposed method was verified by measuring the spacings in reinforced concrete structure, where the twelve locations in the construction site (265 m²) were scanned within 20 seconds per location and a total of 15 minutes was taken. SLAM, consisting of SIFT, RANSAC, and General framework graph optimization algorithms, produces RGB-based 3D and 3D point cloud models, respectively. The minimum resolution of the 3D point cloud was 0.1mm while that of the RGB-based 3D model was 10 mm. Based on the results from both 3D models, the measurement error was from 10.8% to 0.3% in the 3D point cloud and from 28.4% to 3.1% in the RGB-based 3D model. The results demonstrate that the proposed method has great potential for remote structural supervision with respect to its accuracy and objectivity.

• International Journal of Computer Science & Network Security
• /
• v.21 no.11
• /
• pp.345-353
• /
• 2021

Cloud computing represent a new era of computing that's forms through the combination of service-oriented architecture (SOA), Internet and grid computing with virtualization technology. Virtualization is a concept through which every cloud is enable to provide on-demand services to the users. Most IT service provider adopt cloud based services for their users to meet the high demand of computation, as it is most flexible, reliable and scalable technology. Energy based performance tradeoff become the main challenge in cloud computing, as its acceptance and popularity increases day by day. Cloud data centers required a huge amount of power supply to the virtualization of servers for maintain on- demand high computing. High power demand increase the energy cost of service providers as well as it also harm the environment through the emission of CO₂. An optimization of cloud computing based on energy-performance tradeoff is required to obtain the balance between energy saving and QoS (quality of services) policies of cloud. A study about power usage of resources in cloud data centers based on workload assign to them, says that an idle server consume near about 50% of its peak utilization power [1]. Therefore, more number of underutilized servers in any cloud data center is responsible to reduce the energy performance tradeoff. To handle this issue, a lots of research proposed as energy efficient algorithms for minimize the consumption of energy and also maintain the SLA (service level agreement) at a satisfactory level. VM (virtual machine) consolidation is one such technique that ensured about the balance of energy based SLA. In the scope of this paper, we explore reinforcement with fuzzy logic (RFL) for VM consolidation to achieve energy based SLA. In this proposed RFL based active VM consolidation, the primary objective is to manage physical server (PS) nodes in order to avoid over-utilized and under-utilized, and to optimize the placement of VMs. A dynamic threshold (based on RFL) is proposed for over-utilized PS detection. For over-utilized PS, a VM selection policy based on fuzzy logic is proposed, which selects VM for migration to maintain the balance of SLA. Additionally, it incorporate VM placement policy through categorization of non-overutilized servers as- balanced, under-utilized and critical. CloudSim toolkit is used to simulate the proposed work on real-world work load traces of CoMon Project define by PlanetLab. Simulation results shows that the proposed policies is most energy efficient compared to others in terms of reduction in both electricity usage and SLA violation.

• Journal of Broadcast Engineering
• /
• v.27 no.1
• /
• pp.3-12
• /
• 2022

Image deblurring aims to remove image blur, which can be generated while shooting the pictures by the movement of objects, camera shake, blurring of focus, and so forth. With the rise in popularity of smartphones, it is common to carry portable digital cameras daily, so image deblurring techniques have become more significant recently. Originally, image deblurring techniques have been studied using traditional optimization techniques. Then with the recent attention on deep learning, deblurring methods based on convolutional neural networks have been actively proposed. However, most of them have been developed while focusing on better performance. Therefore, it is not easy to use in real situations due to the speed of their algorithms. To tackle this problem, we propose a novel deep learning-based deblurring algorithm that can be operated in real-time on HD resolution. In addition, we improved the training and inference process and could increase the performance of our model without any significant effect on the speed and the speed without any significant effect on the performance. As a result, our algorithm achieves real-time performance by processing 33.74 frames per second at 1280×720 resolution. Furthermore, it shows excellent performance compared to its speed with a PSNR of 29.78 and SSIM of 0.9287 with the GoPro dataset.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.7 no.3
• /
• pp.329-337
• /
• 2017

This paper proposes a search method that explores parameters C and σ values of support vector machines (SVM) to improve performance while maintaining search accuracy. A traditional grid search method requires tremendous computational times because it searches all available combinations of C and σ values to find optimal combinations which provide the best performance of SVM. To address this issue, this paper proposes a deep search method that reduces computational time. In the first stage, it divides C-σ- accurate metrics into four regions, searches a median value of each region, and then selects a point of the highest accurate value as a start point. In the second stage, the selected start points are re-divided into four regions, and then the highest accurate point is assigned as a new search point. In the third stage, after eight points near the search point. are explored and the highest accurate value is assigned as a new search point, corresponding points are divided into four parts and it calculates an accurate value. In the last stage, it is continued until an accurate metric value is the highest compared to the neighborhood point values. If it is not satisfied, it is repeated from the second stage with the input level value. Experimental results using normal and defect bearings show that the proposed deep search algorithm outperforms the conventional algorithms in terms of performance and search time.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.5
• /
• pp.30-42
• /
• 2023

Various studies have been conducted using microtraffic simulation (VISSIM) to analyze the safety of traffic flow when introducing autonomous vehicles. However, no studies have analyzed traffic safety in mixed traffic while considering the driving behavior of general vehicles as a parameter in VISSIM. Therefore, the aim of this study was to optimize the input variables of VISSIM for non-autonomous vehicles through genetic algorithms to obtain realistic behavior. A traffic safety analysis was then performed according to the penetration rate of autonomous vehicles. In a 640 meter section of US highway I-101, the number of conflicts was analyzed when the trailing vehicle was a non-autonomous vehicle. The total number of conflicts increased until the proportion of autonomous vehicles exceeded 20%, and the number of conflicts decreased continuously after exceeding 20%. The number of conflicts between non-autonomous vehicles and autonomous vehicles increased with proportions of autonomous vehicles of up to 60%. However, there was a limitation in that the driving behavior of autonomous vehicles was based on the results of the literature and did not represent actual driving behavior. Therefore, for a more accurate analysis, future studies should reflect the actual driving behavior of autonomous vehicles.

• Smart Media Journal
• /
• v.13 no.4
• /
• pp.33-48
• /
• 2024

With the advent of 5G, characterized by Enhanced Mobile Broadband (eMBB), Ultra-Reliable Low Latency Communications (URLLC), and Massive Machine Type Communications (mMTC), efficient network management and service provision are becoming increasingly critical. This paper proposes a novel approach to address key challenges of 5G networks, namely ultra-high speed, ultra-low latency, and ultra-reliability, while dynamically optimizing network slicing and resource allocation using machine learning (ML) and deep learning (DL) techniques. The proposed methodology utilizes prediction models for network traffic and resource allocation, and employs Federated Learning (FL) techniques to simultaneously optimize network bandwidth, latency, and enhance privacy and security. Specifically, this paper extensively covers the implementation methods of various algorithms and models such as Random Forest and LSTM, thereby presenting methodologies for the automation and intelligence of 5G network operations. Finally, the performance enhancement effects achievable by applying ML and DL to 5G networks are validated through performance evaluation and analysis, and solutions for network slicing and resource management optimization are proposed for various industrial applications.