• 한국전산유체공학회:학술대회논문집
• /
• 1998.05a
• /
• pp.82-89
• /
• 1998

There are many moving-body problems to be solved, the solution of which necessary for proper design of flight vehicles in aerospace industry. Since a body moves relative to other bodies in the category of these problems, difficulty arises regarding both generation of computational grid around the body in motion and conservation of flow properties in the moving grid system. A few example could be store separation from the aircraft and relative vibration of multiple bodies in the high-speed flow passage. In this paper we report on the progress made in computing moving-body aerodynamics related with sabot separation characteristics. Conservative overlapping grid together with cell-merging-unmerging technique is used to solve the Euler equations for a body in high-speed motion. Carbuncle errors has to be removed before we obtain physically adequate solution. Two-dimensional application is reported here.

• Journal of the Korean Society of Visualization
• /
• v.21 no.1
• /
• pp.26-30
• /
• 2023

Interaction of a droplet and substrate is important to determine the coating and final deposition pattern in inkjet printing system. In particular, an accurate deposition of the droplet should be guaranteed for high-resolution patterning. In this study, we performed high-speed shadowgraph experiments on droplet train impact in inkjet system. From the high-speed images, we observed an unexpected bouncing phenomenon. We have found two factors affecting bouncing regime; the Weber number and the curvature of deposited droplet. Experimental results indicate that there is a critical curvature diameter of deposited droplet, which splits into bouncing and merging regime. From this result, we obtained a power-law behavior between the Weber number and the curvature. The understanding of bouncing phenomena helps to improve the accuracy and productivity of inkjet printing.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.2D
• /
• pp.167-173
• /
• 2009

In general, an influenced area of merging section is defined as 500 m including 100 m upstream and 400 m downstream. However, from an observation of the actual traffic flow, it is found that merging effect influences more on downstream than upstream. In this study, an influenced area of merging section on freeway is analyzed by using turbulence which is defined as conflicts between vehicles. In order to overcome the limits of existing traffic flow detection system established with intervals of about 500 m, this study uses raw data collected from the detectors which are established in entrance ramps with similar road conditions. To divide data of each point into similar road conditions, the data of total 72 hours is sorted by Level of Service. An influenced area analyzed by standard deviation of speed is 700 m section of highway, including 300 m upstream and 400 m downstream, for both right and left ramps. It is the result including upstream 200 m more than previous studies.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.40 no.4
• /
• pp.227-236
• /
• 2016

In the present experimental study, the effect of gas- and liquid-injected methods on the formation of bubble and liquid slug at the merging micro T-junction of a square microchannel with dimensions $600{\mu}m{\times}600{\mu}m$ was investigated. Nitrogen and water were used as test fluids. The superficial velocities of the liquid and gas were in the range of 0.05 - 1 m/s, and 0.1 - 1 m/s, respectively, where the Taylor flow was observed. The bubble length, liquid slug length, bubble velocity, and bubble generation frequency were measured by analyzing the images captured using a high-speed camera. Under similar inlet superficial velocity conditions, in the case of gas injection to the main channel at the merging T-junction (T_gas-liquid), the lengths of the bubble and liquid slug were longer, and the bubble generation frequency was lower than in the case of liquid injection to the main channel at the merging T-junction (T_liquid-gas). On the other hand, in both cases, the bubble velocity was almost the same. The previous correlation proposed using experimental data for T_liquid-gas had predicted the present experimental data of bubble length, bubble velocity, liquid slug length, and bubble generation frequency for T_gas-liquid to be ~24%, ~9%, ~39%, ~55%, respectively.

• Journal of Korean Society of Transportation
• /
• v.27 no.4
• /
• pp.115-126
• /
• 2009

The lane-change behavior usually consists of discretionary lane-change and mandatory lane-change types. For the first type, drivers change lanes selectively to maintain their own driving condition and the second type is the case that the drivers must change the current lane, which can occur in recurrent congestion sections like merging and weaving sections. The mandatory lane-change behavior have a great effect on the operation condition of freeway. In this paper, we first generate data such as traffic volumes, speeds, densities, and the number of lane-change within the merging and weaving sections using the data of individual vehicle collected from time-lapse aerial photography. And then, the data is divided into the stable and congested flow by analyzing the speed variation pattern of individual vehicles. In addition, the number of lane-changing from ramp to mainline within every 30-meter interval is investigated before and after traffic congestion at study sites and the distribution of lane-changing at each 30-meter point is analyzed to identify the variation of lane-changing ratio depending on the stable and congested flows. To recognize the effect of mainline flow influenced by ramp flow, this study also analyzes the characteristics of the lane-changing distributions within the lanes of mainline. The purpose of this paper is to present the basic theory to be used in developing a lane-changing model at the merging and weaving sections on freeways.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.8 no.6
• /
• pp.1449-1457
• /
• 2007

A panorama image is constructed by merging several overlapped images to a big one. There are two kinds of methods, feature based and direct method, in the construction. Feature based one has a merit of processing speed faster than direct one. But, it is difficult to be implemented under slower processing environments such as mobile device. This paper proposed high speed construction method of a panorama image. The algorithm extremely improved matching speed by reducing the number of matching parameters using scene shot guider, and additionally adapted local matching technique to reduce matching error caused by the fewer matching parameters. In the experiments, it was shown that the proposed method required about 0.078 second in processing time, about 17 times shorter than the feature based one, for 24-bit color images of $320{\times}240$ size.

• Journal of ILASS-Korea
• /
• v.25 no.3
• /
• pp.103-110
• /
• 2020

In this study, the behavior of water droplet impacting on a thin horizontal wire was visualized by time-delay photography. The impact behavior modes, critical capture speed and trapped mass were analyzed by changing the droplet size, velocity, wire diameter and eccentricity ratio. As the Weber number increased, the hanging, merging, and splitting modes appeared sequentially for the case of central impact, and the hanging and non-splitting modes appeared for the case of off-center impact. The boundary We number of each mode was affected by the diameter ratio. The critical capture speed was affected much by the degree of eccentricity. For all diameter ratios, it was higher for the case of central impact than for off-center impact. The trapped mass was larger for the case of central impact than for off-center impact and it increased with the smaller We number and the larger diameter ratio.

• Journal of Information Processing Systems
• /
• v.19 no.1
• /
• pp.109-117
• /
• 2023

Defect detection is one of the key factors in fabric quality control. To improve the speed and accuracy of denim fabric defect detection, this paper proposes a defect detection algorithm based on cascading feature extraction architecture. Firstly, this paper extracts these weight parameters of the pre-trained VGG16 model on the large dataset ImageNet and uses its portability to train the defect detection classifier and the defect recognition classifier respectively. Secondly, retraining and adjusting partial weight parameters of the convolution layer were retrained and adjusted from of these two training models on the high-definition fabric defect dataset. The last step is merging these two models to get the defect detection algorithm based on cascading architecture. Then there are two comparative experiments between this improved defect detection algorithm and other feature extraction methods, such as VGG16, ResNet-50, and Xception. The results of experiments show that the defect detection accuracy of this defect detection algorithm can reach 94.3% and the speed is also increased by 1-3 percentage points.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.3
• /
• pp.840-860
• /
• 2023

Aiming at the serious occlusion and slow tracking speed in pedestrian target tracking and recognition in complex scenes, a target tracking method based on improved YOLO v5 combined with Deep SORT is proposed. By merging the attention mechanism ECA-Net with the Neck part of the YOLO v5 network, using the CIoU loss function and the method of CIoU non-maximum value suppression, connecting the Deep SORT model using Shuffle Net V2 as the appearance feature extraction network to achieve lightweight and fast speed tracking and the purpose of improving tracking under occlusion. A large number of experiments show that the improved YOLO v5 increases the average precision by 1.3% compared with other algorithms. The improved tracking model, MOTA reaches 54.3% on the MOT17 pedestrian tracking data, and the tracking accuracy is 3.7% higher than the related algorithms and The model presented in this paper improves the FPS by nearly 5 on the fps indicator.

• The Korean Journal of Applied Statistics
• /
• v.22 no.4
• /
• pp.709-721
• /
• 2009

This paper discusses an efficient image segmentation using morphological watershed algorithm that is robust to noise. Morphological image segmentation consists of four steps: image simplification, computation of gradient image and watershed algorithm and region merging. Conventional watershed segmentation exhibits a serious weakness for over-segmentation of images. In this paper we present a morphological edge detection methods for detecting edges under noisy condition and apply our watershed algorithm to the resulting gradient images and merge regions using Kolmogorov-Smirnov test for eliminating irrelevant regions in the resulting segmented images. Experimental results are analyzed in both qualitative analysis through visual inspection and quantitative analysis with percentage error as well as computational time needed to segment images. The proposed algorithm can efficiently improve segmentation accuracy and significantly reduce the speed of computational time.