• Journal of Broadcast Engineering
• /
• v.13 no.6
• /
• pp.915-929
• /
• 2008

In this paper, we propose an algorithm for pixel-based disparity estimation with reliability in the multi-view image. The proposed method estimates an initial disparity map using edge information of an image, and the initial disparity map is used for reducing the search range to estimate the disparity efficiently. Furthermore, disparity-mismatch on object boundaries and textureless-regions get reduced by adaptive block size. We generated intermediate-view images to evaluate the estimated disparity. Test results show that the proposed algorithm obtained $0.1{\sim}1.2dB$ enhanced PSNR(peak signal to noise ratio) compared to conventional block-based and pixel-based disparity estimation methods.

• Proceedings of the KSME Conference
• /
• 2001.11b
• /
• pp.36-41
• /
• 2001

In order to simulate wake of stator and a gas turbine engine's balde row, acryl cylinder and a linear turbine cascade were used respectively in this study. Experimental of heat transfer distributions was done on the passage endwall and blade suction surface. Temperature distributions on the experimental regions were obtained through image processing system by using the cholesteric type liquid crystal which has chain structure of metyl$(CH_3)$. To represent the degree of heat transfer, dimensionless St number was used. The results show that heat transfer on the blade suction surface was increased due to the wake from the cylinder and was decreased as the distance between cylinder row and blade row increases. Because of groth of passage vortex, heat transfer distributions on the trailing edge area showed triangular shape which was little changed with wake. On the other hand, heat transfer on the passage endwall was decreased due to the wake from cylinder. As the distance between cylinder row and blade row increases, heat transfer was more decreased.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.3
• /
• pp.276-282
• /
• 2015

This paper proposes a SMD (surface mounting device) classification method for the PCB assembly inspection machines. The package types of SMD components should be classified to create the job program of the inspection machine. In order to reduce the creation time of job program, we developed the automatic classification algorithm for the SMD packages. We identified the chip-type packages by color and edge distribution of the images. The input images are transformed into the HSI color model, and the binarized histroms are extracted for H and S spaces. Also the edges are extracted from the binarized image, and quantized histograms are obtained for horizontal and vertical direction. The neural network is then applied to classify the package types from the histogram inputs. The experimental results are presented to verify the usefulness of the proposed method.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.26 no.7
• /
• pp.991-996
• /
• 2002

This paper presents the fabrication method of a micro-riblet film (MRF) using MEMS technology and the experimental results of the drag reduction of an airfoil with MRFs. Riblets having grooved surface in the streamwise direction has been proven as an effective passive control technique of the drag reduction. A V-grooved pattern on (100) silicon wafer is etched with anisotropic bulk micromachining. The MRF is completed by replicating the V-grooved pattern with polydimethylsiloxane (PDMS). Experiments were performed by measuring a velocity field behind the trailing edge of a NACA 0012 airfoil with and without MRFs in a closed-type subsonic wind tunnel using particle image velocimetry (PlV) technique. The MRF provides about 3.8 % drag reduction compared to the drag on a smooth airfoil when the freestream velocity of wind tunnel is 3.3 m/s.

• Journal of Korea Multimedia Society
• /
• v.16 no.7
• /
• pp.852-861
• /
• 2013

When users use medical volume rendering applications, selecting specific region of volume data and observing the region by magnification is a common process.As the wood-grain artifact is arise from the magnified image, the jittered sampling technique has been used to remove the problem. However, the jittered sampling leads to some noise along the volume edge. In this research, we reveal the reason of the noise, and present a solution. To remove the wood-grain artifact without the noise, we propose the empty space jittering and the sampling alignment method. Using these methods, we can produce high quality volume rendering images without noticeable time consuming.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.10a
• /
• pp.416-419
• /
• 2003

The cellular neural networks have the circuit structure that differs from the form of general neural network. It consists of an array of the same cell which is a simple processing element, and each of the cells has local connectivity and space invariant template property. In this paper, time-multiplex image processing technique is applied for processing large images using small size CNN cell block, and we simulate the edge detection of a large image using the simulator implemented with a c program and matlab model. A 5$\times$5 CNN hardware and pre post processor is also implemented and is under test.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.10a
• /
• pp.754-757
• /
• 2003

In Oriental medicine, the shape of collarette that formed with position in iris of patients often used by health diagnotcian to grasp health condition. In this paper, we present method that effectively extract collarette that exist in Iris image. After proposed method detert iris area using circular edge detector, derides boundary candidate point through radial line search and threshold value establishment. And boundary candidate line is treated to use nearest neighbor calculation at each boundary candidate point, finally extracts collarette through linear interpolation. As a result of experimenting about iris images, We Confirmed that can be used as assistant tool of diagnostic system that can presume state of ventriculus of human body.

• Journal of IKEEE
• /
• v.24 no.3
• /
• pp.685-690
• /
• 2020

We propose an efficient convolutional neural network with much lower computational complexity and higher accuracy based on MobileNet V2 for mobile or edge devices. The proposed network consists of bottleneck layers with larger expansion factors and adjusted number of channels, and excludes a few layers, and therefore, the computational complexity is reduced by half. The performance the proposed network is verified by measuring the accuracy and execution times by CPU and GPU using ImageNet100 dataset. In addition, the execution time on GPU depends on the CNN architecture.

• Journal of the Korean Society of Combustion
• /
• v.20 no.1
• /
• pp.24-31
• /
• 2015

Flame structure, which contains a useful information for studying combustion instability of the flame, is often quantitatively visualized with PLIF (planar laser-induced fluorescence) and/or chemiluminescence images. The latter, a line-integral of a flame property, needs to be preprocessed before being inverted, mainly due to its inherent noise and the axisymmetry assumption of the inversion. A preprocessing scheme utilizing multi-division of ROI (region of interest) of the chemiluminescence image is proposed. Its feasibility has been tested with OH PLIF and $OH^*$ chemiluminescence images of SNG (synthetic natural gas) swirl-stabilized flames taken from a model gas turbine combustor. It turns out that the multi-division technique outperforms two conventional ones: those are, one without preprocessing and the other with uni-division preprocessing, reconstructing the SNG flame structure much better than its two counterparts, when compared with the corresponding OH PLIF images. It is also found that the Canny edge detection algorithm used for detecting edges in the multi-division method works better than the Sobel algorithm does.

• Korean Journal of Digital Imaging in Medicine
• /
• v.11 no.2
• /
• pp.109-114
• /
• 2009

Carried out an experiment for quantitative evaluation about intraoral sensor among oral dental imaging equipments in PACS environment. For evaluation, testing environment & evaluation criteria are established that refer to the 'IEC 62220-1-2' and experimented with set up the standard radiation penetration that correspond with RQA-3(IEC 61267). Results of experiments using the image J, derived the ESF(edge spread function), LSF (line spread function) and calculated the MTF(modulation transfer function) finally. As a result, the MTF that 0.1, 0.2sec are 10% about 10 lp/mm and 0.32 sec is the 10% about 9 lp/mm in level represents the value of the MTF. Change of scan condition in dental environmental, according to the MTF value taken note that no changes can be seen enough. However, the dimensions of each other size $1(1200{\times}1600)$, size $2(1440{\times}1920)$ intraoral sensors for 3 lp/mm, respectively 40%, 90% of the note might have been the difference between the value of MTF, in accordance with standard sensors might note differences could be observed.