• Advances in nano research
• /
• v.12 no.2
• /
• pp.223-239
• /
• 2022

Object detection has always been to pursue objects with particular properties or representations and to predict details on objects including the positions, sizes and angle of rotation in the current picture. This was a very important subject of computer vision science. While vision-based object tracking strategies for the analysis of competitive videos have been developed, it is still difficult to accurately identify and position a speedy small ball. In this study, deep learning (DP) network was developed to face these obstacles in the study of tennis motion tracking from a complex perspective to understand the performance of athletes. This research has used CNN-VGG 16 to tracking the tennis ball from broadcasting videos while their images are distorted, thin and often invisible not only to identify the image of the ball from a single frame, but also to learn patterns from consecutive frames, then VGG 16 takes images with 640 to 360 sizes to locate the ball and obtain high accuracy in public videos. VGG 16 tests 99.6%, 96.63%, and 99.5%, respectively, of accuracy. In order to avoid overfitting, 9 additional videos and a subset of the previous dataset are partly labelled for the 10-fold cross-validation. The results show that CNN-VGG 16 outperforms the standard approach by a wide margin and provides excellent ball tracking performance.

• The Bulletin of The Korean Astronomical Society
• /
• v.40 no.2
• /
• pp.36.2-36.2
• /
• 2015

We report the first detection of linearly polarized emission at an observing wavelength of 350 mum from the radio-loud active galactic nucleus 3C 279. We conducted polarization observations for 3C 279 using the SHARP polarimeter in the Caltech Submillimeter Observatory on 2014 March 13 and 14. For the first time, we detected the linear polarization with the degree of polarization of $13.3%{\pm}3.4%$ (3.9sigma) and the electric vector position angle (EVPA) of $34.^{\circ}7{\pm}5.^{\circ}6$. We also observed 3C 279 simultaneously at 13, 7, and 3.5 mm in dual polarization with the Korean very long baseline interferometry (VLBI) Network on 2014 March 6 (single dish) and imaged in milliarcsecond (mas) scales at 13, 7, 3.5, and 2.3 mm on March 22 (VLBI). We found that the degree of linear polarization increases from 10% to 13% at 13 mm to 350 mum and the EVPAs at all observing frequencies are parallel within < $10^{\circ}$ to the direction of the jet at mas scale, implying that the integrated magnetic fields are perpendicular to the jet in the innermost regions. We also found that the Faraday rotation measures RM are in a range of $-6.5{\times}102{\sim}-2.7{\times}103$ rad m-2 between 13 and 3.5 mm, and are scaled as a function of wavelength:| {RM}| ${\backslash}propto$ {lambda }-2.2. These results indicate that the millimeter and sub-millimeter polarization emission are generated in the compact jet within 1 mas scale and affected by a Faraday screen in or in the close proximity of the jet.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.1
• /
• pp.107-112
• /
• 2020

Recently, as the field of use of drones is diversified, it is actively used not only for surveying but also for search and rescue work. In these applications it is very important to know the location of the target or the location of the UAV. This paper proposes a target detection method using images taken from drones. The proposed method calculates the latitude and longitude information of the target by finding the location of the target by comparing it with the image to find the image taken by the drone. The exact latitude and longitude information of the target is calculated by calculating the actual distance corresponding to the distance of the image image using the characteristics of the pinhole camera. The proposed method through the actual experiment confirmed that the latitude and longitude of the target was accurately identified.

• Journal of Broadcast Engineering
• /
• v.19 no.3
• /
• pp.396-404
• /
• 2014

The field of searching clothing, which is very difficult due to the nature of the informal sector, has been in an effort to reduce the recognition error and computational complexity. However, there is no concrete examples of the whole progress of learning and recognizing for cloth, and the related technologies are still showing many limitations. In this paper, the whole process including identifying both the person and cloth in an image and analyzing both its color and texture pattern is specifically shown for classification. Especially, deformable search descriptor, LBPROT_35 is proposed for identifying the pattern of clothing. The proposed method is scale and rotation invariant, so we can obtain even higher detection rate even though the scale and angle of the image changes. In addition, the color classifier with the color space quantization is proposed not to loose color similarity. In simulation, we build database by training a total of 810 images from the clothing images on the internet, and test some of them. As a result, the proposed method shows a good performance as it has 94.4% matching rate while the former Dense-SIFT method has 63.9%.

• Tunnel and Underground Space
• /
• v.34 no.1
• /
• pp.1-14
• /
• 2024

An image-reconstruction technology, involving the deployment of an unmanned mobility equipped with high-speed LiDAR (Light Detection And Ranging) has been proposed to reconstruct the shape of abandoned mine. Unmanned mobility operation is remarkably useful in abandoned mines fraught with operational difficulties including, but not limited to, obstacles, sludge, underwater and narrow tunnel with the diameter of 1.5 m or more. For cases of real abandoned mines, quadruped robots, quadcopter drones and underwater drones are respectively deployed on land, air, and water-filled sites. In addition to the advantage of scanning the abandoned mines with 2D solid-state lidar sensors, rotation of radiation at an inclination angle offers an increased efficiency for simultaneous reconstruction of mineshaft shapes and detecting obstacles. Sensor and robot posture were used for computing rotation matrices that helped compute geographical coordinates of the solid-state lidar data. Next, the quadruped robot scanned the actual site to reconstruct tunnel shape. Lastly, the optimal elements necessary to increase utility in actual fields were found and proposed.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.1
• /
• pp.7-15
• /
• 2023

Recently, as interest in air quality in vehicles increases, the use of fine dust detection sensors for air quality measurement is becoming common. An axial-flow fan is inserted in the fine dust sensor installed in the air conditioning system in the vehicle to prevent dust from sinking directly on the sensor. When the sensor operates, the flow noise caused by the rotation of the axial-flow fan acts as a major noise source of the fine dust sensor. flow noise is recognized as one of the product competitiveness of fine dust sensors. In this study, the noise was gradually reduced at the same flow rate by improving the flow performance of the small axial flow fan. First, a virtual fan performance tester consisting of about 20 million grids was developed to analyze the aerodynamic performance of the target small axial-flow fan. In addition, the flow field was simulated by using compressible Large Eddy Simulation for direct computation of flow noise as well as high-accurate prediction of flow rate. The validity of numerical method are confirmed through the comparison of predicted results with experimental ones. After the effects of pitch angle on flow performance were analyzed using the verified numerical method, the pitch angle was determined to maximize the flow rate. It was found that the flow rate was increased by 8.1 % and noise was reduced by 0.8 dBA when the axial-flow fan with the optimum pitch angle was used.

• Proceedings of the KSRS Conference
• /
• v.1
• /
• pp.106-109
• /
• 2006

Over the years, many researches have been performed to create 3D digital maps. Nevertheless, it is still time-consuming and involves a high cost because a large part of 3D digital mapping is conducted manually. To compensate this limitation, we propose methodologies to represent 3D objects as 3D symbols and locate these symbols into a base map automatically. First of all, we constructed the 3D symbol library to represent 3D objects as 3D symbols. In the 3D symbol library, the attribute and geometry information are stored, which defines factors related to the types of symbols and related to the shapes respectively. These factors were used to match 3D objects and 3D symbols. For automatic mapping of 3D symbols into a base map, we used predefined parameters such as the size, the height, the rotation angle and the center of gravity of 3D objects which are extracted from Light Detection and Ranging (LIDAR) data and 2D digital maps. Finally, the 3D map in urban area was constructed and the mapping results were tested using aerial photos as reference data. Through this research, we can identify that the developed the algorithms can be used as effective techniques for 3D digital cartographic techniques

• Journal of the Korean Society for Precision Engineering
• /
• v.30 no.9
• /
• pp.901-908
• /
• 2013

The introduction of industrial intelligent robot using vision sensor has been interested in automated factory. 2D and 3D vision sensors have used to recognize object and to estimate object pose, which is for packaging parts onto a complete whole. But it is not trivial task due to illumination and various types of objects. Object image has distorted due to illumination that has caused low reliability in recognition. In this paper, recognition method of complex shape object has been proposed. An accurate object region has detected from combined binary image, which has achieved using DoG filter and local adaptive binarization. The object has recognized using neural network, which is trained with sub-divided object class according to object type and rotation angle. Predefined shape model of object and maximal slope have used to estimate the pose of object. The performance has evaluated on ETRI database and recognition rate of 96% has obtained.

• Journal of Digital Convergence
• /
• v.11 no.7
• /
• pp.157-163
• /
• 2013

In this paper, we propose a vision and depth information based real-time hand gesture interface method using finger joint estimation. For this, the areas of left and right hands are segmented after mapping of the visual image and depth information image, and labeling and boundary noise removal is performed. Then, the centroid point and rotation angle of each hand area are calculated. Afterwards, a circle is expanded at following pattern from a centroid point of the hand to detect joint points and end points of the finger by obtaining the midway points of the hand boundary crossing and the hand model is recognized. Experimental results that our method enabled fingertip distinction and recognized various hand gestures fast and accurately. As a result of the experiment on various hand poses with the hidden fingers using both hands, the accuracy showed over 90% and the performance indicated over 25 fps. The proposed method can be used as a without contacts input interface in HCI control, education, and game applications.

• Archives of Plastic Surgery
• /
• v.48 no.4
• /
• pp.457-461
• /
• 2021

Background Pedicled perforator flaps can present postoperative complications similar to those encountered in free flap surgery. Beyond a clinical evaluation, there is still no reliable technical aid for the early prediction of vascular issues. The aim of this study was to assess the support of near-infrared spectroscopy technology as an intraoperative tool to anticipate postsurgical flap ischemia. Methods We prospectively enrolled 13 consecutive patients who were referred to our hospital from March 2017 to July 2018 and required a reconstructive procedure with a pedicled fasciocutaneous perforator flap. We measured flap peripheral capillary oxygen saturation (SpO₂) in each patient with a Somanetics INVOS 5100C Cerebral/Somatic Oximeter (Medtronic), both before and after transposition. Patient demographics, operative data, and complications were then recorded during the following 6 months. We analyzed the data using the Wilcoxon signed-rank test and linear regression. Results The mean flap SpO₂ before and after transposition was 92%±3% and 78%±19%, respectively. The mean change in SpO₂ was 14%±17%, with a range of 0% to 55%. The change in saturation and mean saturation ratio were significantly different between patients with and without postoperative flap necrosis. Conclusions An immediate quantitative analysis of flap peripheral capillary SpO₂ after transposition has never before been described. In our experience, an intraoperative drop in SpO₂ equal to or greater than 15%-20% predicted vascular complications in pedicled perforator flaps. Conversely, flap size and rotation angle were not correlated with the risk of flap necrosis.