• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.2
• /
• pp.751-770
• /
• 2019

Action recognition is an essential task in computer vision due to the variety of prospective applications, such as security surveillance, machine learning, and human-computer interaction. The availability of more video data than ever before and the lofty performance of deep convolutional neural networks also make it essential for action recognition in video. Unfortunately, limited crafted video features and the scarcity of benchmark datasets make it challenging to address the multi-person action recognition task in video data. In this work, we propose a deep convolutional neural network-based Effective Hybrid Learning (EHL) framework for two-person interaction classification in video data. Our approach exploits a pre-trained network model (the VGG16 from the University of Oxford Visual Geometry Group) and extends the Faster R-CNN (region-based convolutional neural network a state-of-the-art detector for image classification). We broaden a semi-supervised learning method combined with an active learning method to improve overall performance. Numerous types of two-person interactions exist in the real world, which makes this a challenging task. In our experiment, we consider a limited number of actions, such as hugging, fighting, linking arms, talking, and kidnapping in two environment such simple and complex. We show that our trained model with an active semi-supervised learning architecture gradually improves the performance. In a simple environment using an Intelligent Technology Laboratory (ITLab) dataset from Inha University, performance increased to 95.6% accuracy, and in a complex environment, performance reached 81% accuracy. Our method reduces data-labeling time, compared to supervised learning methods, for the ITLab dataset. We also conduct extensive experiment on Human Action Recognition benchmarks such as UT-Interaction dataset, HMDB51 dataset and obtain better performance than state-of-the-art approaches.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.2
• /
• pp.711-732
• /
• 2019

Many unmanned aerial vehicle (UAV) applications have been employed for performing data collection in facilitating tasks such as surveillance and monitoring objectives in remote and dangerous environments. In light of the fact that most of the existing UAV relaying applications operate in conventional half-duplex (HD) mode, a full-duplex (FD) based UAV relay aided wireless network is investigated, in which the UAV relay helps forwarding information from the source (S) node to the destination (D). Since the activated UAV relays are always floating and flying in the air, its channel state information (CSI) as well as channel capacity is a time-variant parameter. Considering decode-and-forward (DF) relaying protocol in UAV relays, the cooperative relaying channel capacity is constrained by the relatively weaker one (i.e. in terms of signal-to-noise ratio (SNR) or signal-to-interference-plus-noise ratio (SINR)) between S-to-relay and relay-to-D links. The channel capacity can be optimized by adaptively optimizing the transmit power of S and/or UAV relay. Furthermore, a hybrid HD/FD mode is enabled in the proposed UAV relays for adaptively optimizing the channel utilization subject to the instantaneous CSI and/or remaining self-interference (SI) levels. Numerical results show that the channel capacity of the proposed UAV relay aided wireless networks can be maximized by adaptively responding to the influence of various real-time factors.

• KIPS Transactions on Computer and Communication Systems
• /
• v.10 no.1
• /
• pp.1-6
• /
• 2021

Recently, incidents such as proxy surgery (unlicensed medical practice) have been reported in the media that threaten the safety of patients. Alternatives such as the introduction of operating room surveillance camera devices to prevent proxy surgery are emerging, but there are practical difficulties in implementing them due to strong opposition from the medical community. However, the social credibility of doctors is falling as incidents such as proxy surgery occur frequently. In this paper, we propose a medical staff identification system combining Beacon and iris recognition. The system adds reliability by operating on the blockchain network. The system performs primary identification by performing user authentication through iris recognition and proves that the medical staff is in the operating room through beacons. It also ensures patient trust in the surgeon by receiving beacon signals in the background and performing iris authentication at random intervals to prevent medical staff from leaving the operating room after only performing initial certification.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.11
• /
• pp.1437-1444
• /
• 2020

With the development of computer vision systems, many advances have been made in the fields of surveillance, biometrics, medical imaging, and autonomous driving. In the field of autonomous driving, in particular, the object detection technique using deep learning are widely used, and the paved road detection is a particularly crucial problem. Unlike the ROI detection algorithm used in general object detection, the structure of paved road in the image is heterogeneous, so the ROI-based object recognition architecture is not available. In this paper, we propose a deep neural network architecture for atypical paved road detection using Semantic segmentation network. In addition, we introduce the multi-scale semantic segmentation network, which is a network architecture specialized to the paved road detection. We demonstrate that the performance is significantly improved by the proposed method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.1
• /
• pp.15-21
• /
• 2020

In this paper, the observation performance of the electro-optical telescope system which surveils the unknown space objects, is analyzed by the Modeling & Simulation(M&S). The operation concept for the observation of the unknown space objects using two telescope systems is considered and the M&S models are constructed. Based on the operation concept for observing the unknown space objects, the estimated orbit is generated by Initial Orbit Determination(IOD) and the observation performance is analyzed according to the offset compensation cycle for the estimated orbit. The result of the M&S based analysis in this paper shows that the observation performance increases with the shorter offset compensation cycle, and decreases with the longer offset compensation cycle. Therefore, to improve the performance of the telescope system which surveils the unknown space objects, the observation system with accurate initial orbit determination or shorter offset compensation cycle should be designed and constructed.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.4
• /
• pp.173-179
• /
• 2022

Through the 2014 Donbas conflict and the 2022 Ukraine-Russia war, we are experiencing that the command of the air is no longer only secured by the Air Force. Long-range surveillance reconnaissance and strikes carried out by the Air Force could be replaced by drones and missiles, and the enemy's aerial attacks could be controlled by air defense systems such as Panchir and portable anti-aircraft missiles, allowing ground forces to carry out maneuvers freely. In other words, it is much more advantageous for the air force and the navy to take control of the air through long-distance operations, and the ground forces should support them. Therefore, this study aims to consider the cost-effectiveness aspect of the delivery command of the air; it provides implications for quickly responding to enemy air attacks by developing the air defense weapon system, drones, missiles, precision-guided munitions, etc rather than focusing on expensive fighter jets.

• Journal of the Korea Society for Simulation
• /
• v.31 no.2
• /
• pp.11-20
• /
• 2022

There is a critical need for AI assistance in guard operations of Army base perimeters, which is exacerbated by changes in the national defense and security environment such as force reduction. In addition, the possibility for human error inherent to perimeter guard operations attests to the need for an innovative revamp of current systems. The purpose of this study is to propose a real-time object detection AI tailored to military CCTV surveillance with three unique characteristics. First, training data suitable for situations in which relatively small objects must be recognized is used due to the characteristics of military CCTV. Second, we utilize a data augmentation algorithm suited for military context applied in the data preparation step. Third, a noise reduction algorithm is applied to account for military-specific situations, such as camouflaged targets and unfavorable weather conditions. The proposed system has been field-tested in a real-world setting, and its performance has been verified.

• Journal of Environmental Health Sciences
• /
• v.48 no.5
• /
• pp.255-265
• /
• 2022

Background: Research in environmental health (EH) is of crucial strategic importance for contemporary society. It is becoming even more critical in light of the increasingly rapid pace of environmental changes, opportunities, and threats. Objectives: This study aimed to identify trends and the prospective of environmental health research using SWOT analysis. Methods: The trends in environmental health research were reviewed in previous studies and reports. Reviewed manuscripts were searched for using the keywords of 'environmental health' and 'environmental hygiene' in the KCI (Korean Journal of Citation Index), KISS (Korean Academic Information), PubMed, and Google Scholar. Results: It is essential to center the EH research agenda around key priorities focusing on technological innovation, job creation, and the increasingly prominent role of the private sector. Given the rapidly evolving global sustainability agenda, greater clarity on the ever-increasing sources of complexity and growing expectations of the public might be needed. This requires the identification of criteria to identify EH research priorities with the ultimate goal of maximizing societal benefit. Public health relevance, such as extent and severity of health impact, level of exposure, and inequalities of effects, could be included. Conclusions: Considering the recent interest in and importance of environmental health, a comprehensive approach to environmental health research should be required through the application of the latest science and technology, citizen participation, and environmental health surveillance systems.

• Current Optics and Photonics
• /
• v.6 no.5
• /
• pp.445-452
• /
• 2022

We developed an adaptive optics test bench using an optical simulator and two rotating phase plates that mimicked the atmospheric turbulence at Bohyunsan Observatory. The observatory was reported to have a Fried parameter with a mean value of 85 mm and standard deviation of 13 mm, often expressed as 85 ± 13 mm. First, we fabricated several phase plates to generate realistic atmospheric-like turbulence. Then, we selected a pair from among the fabricated phase plates to emulate the atmospheric turbulence at the site. The result was 83 ± 11 mm. To address dynamic behavior, we emulated the atmospheric disturbance produced by a wind flow of 8.3 m/s by controlling the rotational speed of the phase plates. Finally, we investigated how closely the atmospheric disturbance simulation emulated reality with an investigation of the measurements on the optical table. The verification confirmed that the simulator showed a Fried parameter of 87 ± 15 mm as designed, but a little slower wind velocity (7.5 ± 2.5 m/s) than expected. This was because of the nonlinear motion of the phase plates. In conclusion, we successfully mimicked the atmospheric disturbance of Bohyunsan Observatory with an error of less than 10% in terms of Fried parameter and wind velocity.

• Convergence Security Journal
• /
• v.23 no.1
• /
• pp.139-145
• /
• 2023

The ROK Army is promoting cutting-edge, future-oriented military development such as a mobile, intelligent, and hyper-connected Army TIGER system. The future infantry brigade plans to increase mobility with squad-level tactical vehicles to enable combat in multi-domain operations and to deploy various weapon systems such as surveillance and reconnaissance drones. In addition, it will be developed into an intelligent unit that transmits and receives data collected through the weapon system through a hyper-connected network. Accordingly, the future infantry brigade will transmit and receive more data. However, the Army's tactical information communication system has limitations in operating as a tactical communication system for future units, such as low transmission speed and bandwidth and restrictions on communication support. Therefore, in this paper, the information distribution capability of the future infantry brigade is presented through the offensive operation scenario and M&S.