• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.141-148
• /
• 2022

Recently, reinforcement learning combined with deep neural network technology has achieved remarkable success in various fields such as board games such as Go and chess, computer games such as Atari and StartCraft, and robot object manipulation tasks. However, such deep reinforcement learning describes states, actions, and policies in vector representation. Therefore, the existing deep reinforcement learning has some limitations in generality and interpretability of the learned policy, and it is difficult to effectively incorporate domain knowledge into policy learning. On the other hand, dNL-RRL, a new relational reinforcement learning framework proposed to solve these problems, uses a kind of vector representation for sensor input data and lower-level motion control as in the existing deep reinforcement learning. However, for states, actions, and learned policies, It uses a relational representation with logic predicates and rules. In this paper, we present dNL-RRL-based policy learning for transportation mobile robots in a manufacturing environment. In particular, this study proposes a effective method to utilize the prior domain knowledge of human experts to improve the efficiency of relational reinforcement learning. Through various experiments, we demonstrate the performance improvement of the relational reinforcement learning by using domain knowledge as proposed in this paper.

• Journal of Broadcast Engineering
• /
• v.27 no.3
• /
• pp.308-317
• /
• 2022

Researches are underway to efficiently reduce the size of video data transmitted and stored in the image analysis process using deep learning-based machine vision technology. MPEG (Moving Picture Expert Group) has newly established a standardization project called VCM (Video Coding for Machine) and is conducting research on video encoding for machines rather than video encoding for humans. We are researching a multitask that performs various tasks with one image input. The proposed pipeline does not perform all object detection of each task that should precede object detection, but precedes it only once and uses the result as an input for each task. In this paper, we propose a pipeline for efficient multitasking and perform comparative experiments on compression efficiency, execution time, and result accuracy of the input image to check the efficiency. As a result of the experiment, the capacity of the input image decreased by more than 97.5%, while the accuracy of the result decreased slightly, confirming the possibility of efficient multitasking.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.15 no.1
• /
• pp.45-55
• /
• 2022

Currently, the Collective Control Algorithm is the most popular elevator algorithm. The Collective Control Algorithm allows the user to use the elevator when the direction of movement of the elevator and the direction of the user's destination are the same. However, the algorithm has a problem in that only one elevator responds to a user's call when the user's waiting time and using multiple elevators. To solve this problem, this paper proposes a new hybrid HRN algorithm based on the highest response ratio next (HRN) algorithm. In general, HRN Algorithm requires a user's boarding time and getting off time, but due to the nature of the elevator, it is difficult to predict the user's call in advance. Therefore, to overcome these limitations, this paper proposes Hybrid HRN Algorithm that considers the distance between the user's call location and the arrival location. This paper shows that Hybrid HRN Algorithm, proposed through experiments, has an average waiting time of 23.34 seconds, a standard deviation of 11.86, a total moving distance of 535.2m, a total operating time of 84sec, and a driving balance between the two elevators is 92m, which is superior to the previously suggested Collective Control, Zoning, and 3-Passage Algorithm.

• Journal of the Korea Society for Simulation
• /
• v.31 no.3
• /
• pp.1-10
• /
• 2022

Defense fields widely operate unmanned systems to lower vulnerability and enhance combat effectiveness. In the navy, swarm unmanned surface vehicles(USVs) form a cluster within communication range, share situational awareness information among the USVs, and cooperate with them to conduct military missions. This paper proposes an interface system, i.e., Interface Adapter System(IAS), to achieve inter-USV and intra-USV interoperability. We focus on the mission planning subsystem(MPS) for interoperability, which is the core subsystem of the USV to decide courses of action such as automatic path generation and weapon assignments. The central role of the proposed system is to exchange interface data between MPSs and other subsystems in real-time. To this end, we analyzed the operational requirements of the MPS and identified interface messages. Then we developed the IAS using the distributed real-time middleware. As experiments, we conducted several integration tests at swarm USVs simulation environment and measured delay time and loss ratio of interface messages. We expect that the proposed IAS successfully provides bridge roles between the mission planning system and other subsystems.

• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.10
• /
• pp.373-380
• /
• 2022

This paper presents a deep learning based group synchronization that supports networked immersive interactions between remote users. The goal of group synchronization is to enable all participants to synchronously interact with others for increasing user presence Most previous methods focus on NTP-based clock synchronization to enhance time accuracy. Moving average filters are used to control media playout time on the synchronization server. As an example, the exponentially weighted moving average(EWMA) would be able to track and estimate accurate playout time if the changes in input data are not significant. However it needs more time to be stable for any given change over time due to codec and system loads or fluctuations in network status. To tackle this problem, this work proposes the Deep Group Synchronization(DeepGroupSync), a group synchronization based on deep learning that models important features from the data. This model consists of two Gated Recurrent Unit(GRU) layers and one fully-connected layer, which predicts an optimal playout time by utilizing the sequential playout delays. The experiments are conducted with an existing method that uses the EWMA and the proposed method that uses the DeepGroupSync. The results show that the proposed method are more robust against unpredictable or rapid network condition changes than the existing method.

• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.10
• /
• pp.363-372
• /
• 2022

Since the recent COVID-19 Pandemic, the ransomware fandom has intensified along with the expansion of remote work. Currently, anti-virus vaccine companies are trying to respond to ransomware, but traditional file signature-based static analysis can be neutralized in the face of diversification, obfuscation, variants, or the emergence of new ransomware. Various studies are being conducted for such ransomware detection, and detection studies using signature-based static analysis and behavior-based dynamic analysis can be seen as the main research type at present. In this paper, the frequency of ".text Section" Opcode and the Native API used in practice was extracted, and the association between feature information selected using K-means Clustering algorithm, Cosine Similarity, and Pearson correlation coefficient was analyzed. In addition, Through experiments to classify and detect worms among other malware types and Cerber-type ransomware, it was verified that the selected feature information was specialized in detecting specific ransomware (Cerber). As a result of combining the finally selected feature information through the above verification and applying it to machine learning and performing hyper parameter optimization, the detection rate was up to 93.3%.

• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.11
• /
• pp.381-394
• /
• 2022

Recently, due to the personalization and customization of data, Internet-based services have increased requirements for real-time processing, such as real-time AI inference and data analysis, which must be handled immediately according to the user's situation or requirement. Real-time tasks have a set deadline from the start of each task to the return of the results, and the guarantee of the deadline is directly linked to the quality of the services. However, traditional container systems are limited in operating real-time tasks because they do not provide the ability to allocate and manage deadlines for tasks executed in containers. In addition, tasks such as AI inference and data analysis basically utilize graphical processing units (GPU), which typically have performance impacts on each other because performance isolation is not provided between containers. And the resource usage of the node alone cannot determine the deadline guarantee rate of each container or whether to deploy a new real-time container. In this paper, we propose a monitoring technique for tracking and managing the execution status of deadlines and real-time GPU tasks in containers to support real-time processing of GPU tasks running on containers, and a node list management technique for container placement on appropriate nodes to ensure deadlines. Furthermore, we demonstrate from experiments that the proposed technique has a very small impact on the system.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.1-16
• /
• 2022

Despite recent breakthroughs in deep learning and computer vision fields, the pixel-wise identification of tiny objects in high-resolution images with complex disturbances remains challenging. This study proposes a modified U-net for tiny crack segmentation in real-world steel-box-girder bridges. The modified U-net adopts the common U-net framework and a novel Self-Attention-Self-Adaption (SASA) neuron as the fundamental computing element. The Self-Attention module applies softmax and gate operations to obtain the attention vector. It enables the neuron to focus on the most significant receptive fields when processing large-scale feature maps. The Self-Adaption module consists of a multiplayer perceptron subnet and achieves deeper feature extraction inside a single neuron. For data augmentation, a grid-based crack random elastic deformation (CRED) algorithm is designed to enrich the diversities and irregular shapes of distributed cracks. Grid-based uniform control nodes are first set on both input images and binary labels, random offsets are then employed on these control nodes, and bilinear interpolation is performed for the rest pixels. The proposed SASA neuron and CRED algorithm are simultaneously deployed to train the modified U-net. 200 raw images with a high resolution of 4928 × 3264 are collected, 160 for training and the rest 40 for the test. 512 × 512 patches are generated from the original images by a sliding window with an overlap of 256 as inputs. Results show that the average IoU between the recognized and ground-truth cracks reaches 0.409, which is 29.8% higher than the regular U-net. A five-fold cross-validation study is performed to verify that the proposed method is robust to different training and test images. Ablation experiments further demonstrate the effectiveness of the proposed SASA neuron and CRED algorithm. Promotions of the average IoU individually utilizing the SASA and CRED module add up to the final promotion of the full model, indicating that the SASA and CRED modules contribute to the different stages of model and data in the training process.

• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.9
• /
• pp.305-316
• /
• 2022

With the flood of digital data owing to the Internet of Things and big data, cloud service providers that process and store vast amount of data from multiple users can apply duplicate data elimination technique for efficient data management. The user experience can be improved as the notion of edge computing paradigm is introduced as an extension of the cloud computing to improve problems such as network congestion to a central cloud server and reduced computational efficiency. However, the addition of a new edge device that is not entirely reliable in the edge computing may cause increase in the computational complexity for additional cryptographic operations to preserve data privacy in duplicate identification and elimination process. In this paper, we propose an efficiency-improved duplicate data elimination protocol while preserving data privacy with an optimized user-edge-cloud communication framework by utilizing a trusted execution environment. Direct sharing of secret information between the user and the central cloud server can minimize the computational complexity in edge devices and enables the use of efficient encryption algorithms at the side of cloud service providers. Users also improve the user experience by offloading data to edge devices, enabling duplicate elimination and independent activity. Through experiments, efficiency of the proposed scheme has been analyzed such as up to 78x improvements in computation during data outsourcing process compared to the previous study which does not exploit trusted execution environment in edge computing architecture.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.21 no.5
• /
• pp.274-286
• /
• 2022

Multi-object tracking has been studied for a long time under computer vision and plays a critical role in applications such as autonomous driving and driving assistance. Multi-object tracking techniques generally consist of a detector that detects objects and a tracker that tracks the detected objects. Various publicly available datasets allow us to train a detector model without much effort. However, there are relatively few publicly available datasets for training a tracker model, and configuring own tracker datasets takes a long time compared to configuring detector datasets. Hence, the detector is often developed separately with a tracker module. However, the separated tracker should be adjusted whenever the former detector model is changed. This study proposes a system that can train a model that performs detection and tracking simultaneously using only the detector training datasets. In particular, a Siam network with augmentation is used to compose the detector and tracker. Experiments are conducted on public datasets to verify that the proposed algorithm can formulate a real-time multi-object tracker comparable to the state-of-the-art tracker models.