• 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.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.2
• /
• pp.7-14
• /
• 2021

By changing the bulky monolithic services architecture to a microservices-based architecture, industries are managing the rising complexity of Autonomous Vehicles. However, the underlying communication mechanisms for the utilization and distribution of these microservices are incapable of fulfilling the requirements of the futuristic AV, because of the stringent latency requirements along with intermittent and short-lived connectivity issues. This paper proposes to tackle these challenges by employing the revolutionary information-centric networking (ICN) paradigm as an underlying communication architecture. This paper argues that a microservice approach to building autonomous vehicle systems should utilize ICN to achieve effective service utilization, efficient distribution, and uniform service discovery. This research claims that the vision of an information-centric microservices will help to focus on research that can fill in current communication gaps preventing more effective, and lightweight autonomous vehicle services and communication protocols.

• Smart Structures and Systems
• /
• v.24 no.5
• /
• pp.669-681
• /
• 2019

Bridge collapses may deliver a huge impact on our society in a very negative way. Out of many reasons why bridges collapse, poor maintenance is becoming a main contributing factor to many recent collapses. Furthermore, the aging of bridges is able to make the situation much worse. In order to prevent this unwanted event, it is indispensable to conduct continuous bridge monitoring and timely maintenance. Visual inspection is the most widely used method, but it is heavily dependent on the experience of the inspectors. It is also time-consuming, labor-intensive, costly, disruptive, and even unsafe for the inspectors. In order to address its limitations, in recent years increasing interests have been paid to the use of unmanned aerial vehicles (UAVs), which is expected to make the inspection process safer, faster and more cost-effective. In addition, it can cover the area where it is too hard to reach by inspectors. However, this strategy is still in a primitive stage because there are many things to be addressed for real implementation. In this paper, a typical procedure of bridge inspection using UAVs consisting of three phases (i.e., pre-inspection, inspection, and post-inspection phases) and the detailed tasks by phase are described. Also, three major challenges, which are related to a UAV's flight, image data acquisition, and damage identification, respectively, are identified from a practical perspective (e.g., localization of a UAV under the bridge, high-quality image capture, etc.) and their possible solutions are discussed by examining recently developed or currently developing techniques such as the graph-based localization algorithm, and the image quality assessment and enhancement strategy. In particular, deep learning based algorithms such as R-CNN and Mask R-CNN for classifying, localizing and quantifying several damage types (e.g., cracks, corrosion, spalling, efflorescence, etc.) in an automatic manner are discussed. This strategy is based on a huge amount of image data obtained from unmanned inspection equipment consisting of the UAV and imaging devices (vision and IR cameras).

• Electronics and Telecommunications Trends
• /
• v.36 no.1
• /
• pp.89-98
• /
• 2021

The Ministry of National Defense aims to create an environment in which transparent and reasonable defense policies can be implemented in real time by establishing the vision of smart defense innovation based on the Fourth Industrial Revolution and promoting innovation in technology-based defense operation systems. Artificial intelligence (AI) based defense technology is at the level of basic research worldwide, includes no domestic tasks, and involves classified military operation data and command control/decision information. Further, it is needed to secure independent technologies specialized for our military. In the army, military power continues to decline due to aging and declining population. In addition, it is expected that there will be more than 500,000 units should be managed simultaneously, to recognize the battle situation in real time on the future battlefields. Such a complex battlefield, command decisions will be limited by the experience and expertise of individual commanders. Accordingly, the study of AI core technologies supporting real-time combat command is actively pursued at home and abroad. It is necessary to strengthen future defense capabilities by identifying potential threats that commanders are likely to miss, improving the viability of the combat system, ensuring smart commanders always win conflicts and providing reasonable AI digital staff based on data science. This paper describes the recent research trends in AI military staff technology supporting commander decision-making, broken down into five key areas.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.3
• /
• pp.201-210
• /
• 2021

This paper introduces the futsal video analysis system consisting of an analysis program using object tracking technology and a web server that visualizes and provides analyzed data. In this paper, small and medium-sized organizations and amateur players are unable to provide game analysis services, so they propose a system that can solve this problem through this paper. Existing analytical systems use special devices or high-cost cameras, making them difficult for users to use. Thus, in this paper, a system is designed and developed to analyze the competitors' competitions and visualize the data using flat images only. Track an object and calculate the accumulated values to obtain the distance per pixel of the object and extract speed-related data and distance-based data based on it. Converts extracted data to graphs and images through a visualization library, making it convenient to use through web pages. Through this analysis system, we improve the problems of the existing analysis system and make data-based scientific and efficient analysis available.

• KIPS Transactions on Computer and Communication Systems
• /
• v.10 no.1
• /
• pp.7-14
• /
• 2021

In jewelry industry, the proportion of labor costs is high. Also, the production time and quality of products are highly varied depending on the workers' capabilities. Therefore, there is a demand from the jewelry industry for automation. The ear pin insertion automation system is the robot automatically inserts the ear pins into the silicone mold, and this automated system require accurate and fast hole detection method. In this paper, we propose optimal binarization method and a template matching method that can be applied in the ear pin insertion automation system. Through the performance test, it was shown that the applied method has an accuracy of 98.5% and 0.5 seconds faster processing speed than the Otsu binarization method. So, this automation system can contribute to cost reduction, work time reduction, and productivity improvement.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.34 no.5
• /
• pp.215-221
• /
• 2020

Adulterated herbal medicine is intentionally added with undeclared improper or inferior ingredients which should not be in herbal medicine. The contamination with potentially hazardous substances such as heavy metal, pesticides, fungus, and microorganism sometimes can be regarded as one of adulteration in a broad sense. The problem of adulteration is that adulterated herbal medicine shows poor quality and/or can cause adverse events. Therefore, it is important to control adulteration issues for quality assurance and qualitative improvement of herbal medicines. This study aims to summarize and make a reference how to control adulterated herbal medicine. In this process, this study is to investigate studies about adulterated herbal medicine via searching Korean and foreign electronic databases such as PubMed, NDSL and OASIS. Finally eighteen papers were included to this study and analyzed according to the type of study, the category and efficacy of adulterants, the type of analysis methodologies and possible adverse events of adulterants. Phosphodiesterase type 5 (PDE-5) inhibitors for male sexual enhancement and anorexic, laxative, diuretic agents for weight loss and treating obesity has been used frequently as adulterants. The range of adverse event caused by adulterated herbal medicine were very wide from mild symptoms such as diarrhea, constipation, dizziness and blurred vision to very severe symptoms such as heart failure, hypoglycemia and renal impairment. This study showed the recent trend on the research of adulterated herbal medicine and this will be the ground to develop more detailed systems to control adulterated herbal medicine.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.3
• /
• pp.396-402
• /
• 2021

As the world is trying to establish smart city, unmanned vehicle control systems are being widely used. This paper writes about an unmanned parking control system that uses an adaptive background image modeling method, suggesting the method of updating the background image, modeled with an adaptive Gaussian mixture model, in both global and local way according to the moving object. Specifically, this paper focuses on suggesting two methods; a method of minimizing the influence of a moving object on a background image and a method of accurately updating the background image by quickly removing afterimages of moving objects within the area of interest to be monitored. In this paper, through the implementation of the unmanned vehicle control system, we proved that the proposed system can quickly and accurately distinguish both moving and static objects such as vehicles from the background image.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.6
• /
• pp.813-826
• /
• 2022

AI-based image processing technologies in various fields have been widely studied. However, the lighter the board, the more difficult it is to reduce the weight of image processing algorithm due to a lot of computation. In this paper, we propose a method using deep learning for object recognition algorithm in lightweight embedded boards. We can determine the area using a deep neural network architecture algorithm that processes semantic segmentation with a relatively small amount of computation. After masking the area, by using more accurate deep learning algorithm we could operate object detection with improved accuracy for efficient neural network (ENet) and You Only Look Once (YOLO) toward executing object recognition in real time for lightweighted embedded boards. This research is expected to be used for autonomous driving applications, which have to be much lighter and cheaper than the existing approaches used for object recognition.

• Journal of the Institute of Convergence Signal Processing
• /
• v.22 no.4
• /
• pp.156-163
• /
• 2021

Attention mechanisms have been used in deep learning-based computer vision systems, including single image super-resolution (SISR) networks. However, existing SISR networks with attention mechanism focused on real image super-resolution, so it is hard to know whether they are available for CG or VR images. In this paper, we attempt to apply a recent attention module, called balanced attention mechanism (BAM) module, to 12 state-of-the-art SISR networks, and then check whether the BAM module can achieve performance improvement in CG or VR image super-resolution. In our experiments, it has been confirmed that the performance improvement in CG or VR image super-resolution is limited and depends on data characteristics, size, and network type.