• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.12
• /
• pp.561-568
• /
• 2021

In recent years, autonomous driving technologies have become a high-value-added technology that attracts attention in the fields of science and industry. For smooth Self-driving, it is necessary to accurately detect an object and estimate its movement speed in real time. CNN-based deep learning algorithms and conventional dense optical flows have a large consumption time, making it difficult to detect objects and estimate its movement speed in real time. In this paper, using a single camera image, fast object detection was performed using the YOLOv5 algorithm, a deep learning algorithm, and fast estimation of the speed of the object was performed by using a local dense optical flow modified from the existing dense optical flow based on the detected object. Based on this algorithm, we present a system that can predict the collision time and probability, and through this system, we intend to contribute to prevent tram accidents.

• 한국전산유체공학회:학술대회논문집
• /
• 2010.05a
• /
• pp.412-415
• /
• 2010

Research about pedestrian flow in the view of sociology and psychology has been studied for more than a few decade. Due to the advance of computational facility, computational study for pedestrian flow extended to the field of architecture and traffic engineering. However, there is few study for the extremely high dense condition where pedestrian flow is driven by contact force among pedestrian. In this research, we analyze highly dense pedestrian flow using discrete element method

• Journal of Mechanical Science and Technology
• /
• v.19 no.4
• /
• pp.1027-1035
• /
• 2005

Thermophoresis in dense gases is studied by using a multi-scale approach and Born- Yvon­Green (BYG) equation. The problem of a particle movement in an ambient dense gas under temperature gradient is divided into inter and outer ones. The pressure gradient in the inner region is obtained from the solutions of BYG equation. The velocity profile is derived from the conservation equations and calculated using the pressure gradient, which provides the particle velocity in the outer problem. It is shown that the temperature gradient applied to the quiescent ambient gas induces some pressure gradient and thus flow tangential to the particle surface in the interfacial region. The mechanism that induces the flow may be the dominant source of the thermophretic particle movement in dense gases. It is also shown that the particle velocity has a nonlinear relationship with the applied temperature gradient and decreases with increasing temperature.

• Journal of Korea Multimedia Society
• /
• v.21 no.4
• /
• pp.478-484
• /
• 2018

The choice of the motion features influences the result of the human action recognition method directly. Many factors often influence the single feature differently, such as appearance of the human body, environment and video camera. So the accuracy of action recognition is restricted. On the bases of studying the representation and recognition of human actions, and giving fully consideration to the advantages and disadvantages of different features, the Dense Trajectories(DT) algorithm is a very classic algorithm in the field of behavior recognition feature extraction, but there are some defects in the use of optical flow images. In this paper, we will use the improved Dense Trajectories(iDT) algorithm to optimize and extract the optical flow features in the movement of human action, then we will combined with Support Vector Machine methods to identify human behavior, and use the image in the KTH database for training and testing.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.215-216
• /
• 2020

Optical Flow는 컴퓨터 비전 분야의 많은 응용기술에 사용된다. 객체 탐지, 추적, 연속 영상 보간, 3D Reconstruction과 같은 최근에 활발히 연구되는 여러 분야에서 사용되는 기반 기술이다. 최근 딥러닝을 기반으로 한 다양한 연구가 활발히 진행되어 왔으며 높은 정확도를 보이고 있다. 이런 분야들은 많은 경우에 실시간 시스템에 적용되어 이미지로부터 정보를 연산한다. 본 논문은 MaskFlownet, SelFlow, LiteFlowNet2 등과 같은 높은 정확도를 가진 신경망 네트워크로 추정된 Optical Flow를 살펴본다. 각 신경망 네트워크로 얻어진 정확도를 비교하고 디스플레이 기술과 이미지 센서 기술의 발전으로 사용 수요가 많아진 고화질의 이미지를 실시간으로 처리하는 경우, 적용 가능한 Optical Flow의 성능을 분석하였다.

• Korean Chemical Engineering Research
• /
• v.44 no.5
• /
• pp.468-475
• /
• 2006

For the filtration of super compactible cake, the high filtration pressure can not improve filtration rate. As the high pressure, in this case, decreases the cake porosity adjacent to filter medium and thus forms 'dense skin' which decreases the rate of liquid flow in a great extent. Actually, there was no method to improve filtration rate for the filtration with super compactible cake. We propose the saturation of $CO_2$ gas into the suspension before the filtration operation for improving the filtration rate. The dissolved $CO_2$ gas transforms itself into gas phase in the dense skin through which the pressure changes dramatically. The gas secures its space inside the dense skin, and finally forms the flow passages which improve the filtration rate.

• Proceedings of the Korean Society of Soil and Groundwater Environment Conference
• /
• 2003.04a
• /
• pp.51-54
• /
• 2003

Three kinds of experiments were conducted to test existing methods and develop an effective methodology for the remediation of DNAPL trapped in vertical dead-end fractures. A water-flushing method failed to remove TCE from vertical dead-end fractures where no fluid flow occurs. A water-flushing experiment implies that existing remediation methods, utilizing water-based remedial fluid such as surfactant-enhanced method, have difficulty in removing DNAPL trapped from the vertical downward dead-end fractures, because of no water flow through dead-end fractures, capillary, and gravity forces. Fluid denser than TCE was injected into the fracture network, but did not displace TCE from the vertical dead-end fractures. Base(B on the analysis of the experiments, the increase in the density of the dense fluid and the addition of surfactant to the dense fluid were suggested, and this composite dense fluid with surfactant effectively removed TCE from the vertical dead-end fractures.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.9
• /
• pp.4108-4122
• /
• 2016

To address challenges of an unprecedented growth in mobile data traffic, the ultra-dense network deployment is a cost efficient solution to off-load the traffic over other small cells. However, the real traffic is often much lower than the peak-hour traffic and certain small cells are superfluous, which will not only introduce extra energy consumption, but also impose extra interference onto the radio environment. In this paper, an elastic energy efficient cell management scheme is proposed based on flow scheduling among multi-layer ultra-dense cells by a SDN controller. A significant power saving was achieved by a cell-level energy manager. The scheme is elastic for energy saving, adaptive to the dynamic traffic distribution in the office or campus environment. In the end, the performance is evaluated and demonstrated. The results show substantial improvements over the conventional method in terms of the number of active BSs, the handover times, and the switches of BSs.

• Journal of ILASS-Korea
• /
• v.22 no.1
• /
• pp.13-21
• /
• 2017

The fog screen is a device projecting the media to the aerosol flow field. As major parameters to generate dense and steady fog screen, shear stress, optical blockage ratio and SMD were obtained result through experiment. The micro droplet was generated by the piezo oscillation element, and the aerosol flow mixed with an air flow was sprayed into the vertical direction from the top of the fog screen through the 280 mm slot. For produce a dense, uniform fog screen, the shear effect, optical blockage ratio and SMD between aerosol and air curtain were measured. The minimum and maximum shear stress conditions were selected and it was confirmed that the optical transmission deviation of the aerosol flow field was small when the aerosol and air curtain flow rates were changed. When the aerosol and air curtain flow power were 18 V (1.51 m/s) and 24 V (2.55 m/s), respectively, under the condition of the minimum shear stress and laminar flow, the optical blockage ratios with the spray length were small, and it produced a most stable and high density uniform fog screen by injecting a constant of $10{\mu}m$ or less.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.11 no.5
• /
• pp.277-285
• /
• 2016

Moving object detection system has been an emerging research field in various advanced driver assistance systems (ADAS) and surveillance system. In this paper, we propose two optical flow based moving object detection methods at dynamic scenes. Both proposed methods consist of three successive steps; pre-processing, foreground segmentation, and post-processing steps. Two proposed methods have the same pre-processing and post-processing steps, but different foreground segmentation step. Pre-processing calculates mainly optical flow map of which each pixel has the amplitude of motion vector. Dense optical flows are estimated by using Farneback technique, and the amplitude of the motion normalized into the range from 0 to 255 is assigned to each pixel of optical flow map. In the foreground segmentation step, moving object and background are classified by using the optical flow map. Here, we proposed two algorithms. One is Gaussian mixture model (GMM) based background subtraction, which is applied on optical map. Another is adaptive thresholding based foreground segmentation, which classifies each pixel into object and background by updating threshold value column by column. Through the simulations, we show that both optical flow based methods can achieve good enough object detection performances in dynamic scenes.