• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2002.10a
• /
• pp.1049-1052
• /
• 2002

The aligning between the punch and die governs no only the burr formation characteristics but also the life time of the punch and die in the sheet metal blanking process. There are many ways to adjust the two elements in the general punching systems but in the case of micro punch system, the punch size is reduced to a few tenth of micrometer range and the general aligning methods are almost impossible to apply. The image processing is the most widely used method in micro punch aligning, but in order to apply the method, it needs quite a large space for visionary system to approach the punch-die aligning zone. In this paper, the new punch-die aligning method with using the total capacitance between the punch and die hole is proposed. In this method, the tip surface of the punch tool locates at the same plane of the die surface and the capacitance variation between the two elements are measured. When the center of the two elements are coincided, the capacitance is minimized, but when the align Is changed to any direction, the capacitance between the two elements increase. In order to verify the feasibility of this method, the aligning and punching tests was performed.

• ETRI Journal
• /
• v.38 no.5
• /
• pp.847-857
• /
• 2016

Broadcasting services such as multi/single channel HDTV and 3DTV/2DTV use a multi-channel encoder that changes the bitrate and composition of the video service depending on the time. However, this type of multi-channel encoder could cause a longer latency owing to the variable bitrate and relatively bigger size of the buffers, which results in the same delay as in 3DTV even for a conventional DTV service. On the other hand, systems built based on separate encoders, each of which is optimized for the target service, might not have such latency problems. Nevertheless, there might be a distortion problem in the image and sound at the time of a switchover between two encoders with different output bitrates and group of picture structures. This paper proposes a system that can realize a seamless video service conversion using two different video encoders optimized for each video service. An overall functional description of the video service change control server, which is a main control block for the proposed system, is also provided. The experiment results confirm the seamless switchover and reduced broadcasting latency of DTV services compared with a broadcasting system composed of a multi-channel encoder system.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.32 no.8
• /
• pp.574-581
• /
• 2008

The characteristics of combustion and flow for a lean premixed flame in lab-scale gas turbine combustor was studied through experiment and numerical analysis. From the experiment, flame structure and heat release rate were obtained from OH emission spectroscopy. Qualitative comparisons were made line-integrated OH chemiluminescence image and abel-transformed one. NOx analyzer was implemented to get the characteristic of NOx exhaust from the combustor. From the numerical analysis, the thermal distribution and characteristic of recirculation zone with the change of fuel-air mixing degree, the characteristic of methane distribution with equivalence ratio in the combustor respectively. Total heat release rate is increased with increasing equivalence ratio. Thermal Nox is reduced with increasing fuel-air mixing degree. Increasing equivalence ratio results in the decrease of the size of reaction zone and alteration of the position of the reaction zone into the entrance of the combustor.

• Journal of IKEEE
• /
• v.16 no.4
• /
• pp.382-388
• /
• 2012

An H.264/AVC fast motion estimation methodology is presented in this paper. Advanced SIMD based NEON which is one of the parallel processing methods is supported under the ARM Cortex-A9 dual-core platform. NEON is applied to a full search technique with one of the various motion estimation methods and SAD operation count of each macroblock is reduced to 1/4. Pixel values of the corresponding macroblock are assigned to eight 16-bit NEON registers and Intrinsic function in NEON architecture carried out 128 bits arithmetic operations at the same time. In this way, the exact motion vector with the minimum SAD value among the calculated SAD values can be designated. Experimental results show that performance gets improved 30% above average in accordance with the size of image and macroblock.

• International Journal of Internet, Broadcasting and Communication
• /
• v.7 no.1
• /
• pp.10-14
• /
• 2015

Video Content owners seek to squeeze the maximum amount of revenue from their assets via distribution into more territories. Digital Cinema Package(DCP), trusted solution to distribute protected content to theaters, caters to relatively small user base, which limits revenue. With the growth of the Internet & other digital media, the economics of media content has changed dramatically. Security remains main concern to deliver content to millions of consumers using intelligent digital display devices like Tablets, Smartphones, Smart TVs, Desktop & Laptop. By making the video content available to this segment securely, content owners will benefit from increased revenue. Through this paper we propose Open Source HomeDCP model to distribute the content to home users for offline viewing. We propose to include other open source CODEC than JPEG2000/MPEG2, which are specifically designed for theatrical performance. Final image size will be further reduced considering the display device resolution where video will be finally played. Key Delivery Message(KDM) system to be altered to suit new devices. This will be a big boost to Content Economy as content owners would be able to distribute the content securely to the wider audience & ensure more revenue.

• Korean Journal of Metals and Materials
• /
• v.49 no.8
• /
• pp.619-627
• /
• 2011

The poisoning effect of Zr in aluminum alloys was investigated by analyzing the filtered cakes of aluminum alloy melt taken with the $Prefil^{(R)}$ footprinter through a variety of analytic instruments, SEM/EDX, Auger, and TEM. Experimental results indicated that the morphology and chemical composition of the aluminum alloys were not modified with the addition of Zr, which is to previous belief that Zr poisoning is caused by modification of $(Ti_{1-x}Zr_x)Al_3$. On the other hand, $TiAl_3$ surroundig $TiB_2$ particles was modified and its lattice parameter was more mismatched by increasing Zr content, leading to less nucleation rate. This is also supported by the observation that the poisoning effect is reduced when Ti is added, resulting in a lower content ratio of Zr to Ti. These results suggest that extra Ti should be added to eliminate the poisoning effect of Zr in aluminum alloys containing Zr.

• Journal of Information Processing Systems
• /
• v.16 no.6
• /
• pp.1293-1308
• /
• 2020

In fine-tuning-based transfer learning, the size of the dataset may affect learning accuracy. When a dataset scale is small, fine-tuning-based transfer-learning methods use high computing costs, similar to a large-scale dataset. We propose a mid-level feature extractor that retrains only the mid-level convolutional layers, resulting in increased efficiency and reduced computing costs. This mid-level feature extractor is likely to provide an effective alternative in training a small-scale medical image dataset. The performance of the mid-level feature extractor is compared with the performance of low- and high-level feature extractors, as well as the fine-tuning method. First, the mid-level feature extractor takes a shorter time to converge than other methods do. Second, it shows good accuracy in validation loss evaluation. Third, it obtains an area under the ROC curve (AUC) of 0.87 in an untrained test dataset that is very different from the training dataset. Fourth, it extracts more clear feature maps about shape and part of the chest in the X-ray than fine-tuning method.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.1
• /
• pp.181-187
• /
• 2024

There are two kinds of well-known CNN methods, the group equivariant CNN and the CNN using steerable filters, which have excellent classification performances for randomly rotated objects in image space. This paper describes their mathematical structures and introduces implementation methods. We implement them, including the existing CNN, which have the same number of filters, then compare and analyze their performances by simulating them with the randomly rotated MNIST. According to the experimental results, the steerable CNN, which shows a classification improvement over the others, has a relatively small number of parameters to learn, so performance degradation is relatively small even when the size of the training dataset is reduced.

• Journal of Korea Multimedia Society
• /
• v.9 no.4
• /
• pp.434-442
• /
• 2006

This paper proposes a color-matching 3D look-up table that simplifies the complex color-matching procedure between a monitor and a mobile display device, where the image colors are processed in a device-independent color space, such as CIEXYZ or CIELAB, and gamut mapping performed to compensate the gamut difference. The transform from a device-dependent RGB color space to a device-independent color space is implemented by performing display characterization. The mobile LCD characterization error using the S-curve model is larger than the tolerance error since the mobile LCD has the channel-chromaticity-inconstancy and channel-dependence characteristics. In this paper we reduced the characterization error using the electro-optical transfer functions of X, Y, and Z value for R, G, B, C, M, Y, K components. Experimental results demonstrated that 64 ($4{\times}4{\times}4$) was the smallest size of color-matching look-up table that could produce an image with an acceptable reproduction error, based on a comparison of color-matched images resulting from the proposed color-matching look-up table and complex step-by-step color-matching procedures.

• Journal of Internet Computing and Services
• /
• v.18 no.1
• /
• pp.21-26
• /
• 2017

In this paper, an USB-OTG (Uiversal Serial Bus On-the-go) interface module was designed with the iris information for personal identification. The image recognition algorithm which was searching face and iris areas, was proposed with pixel differences from eye blinking after several facial images were captured and then detected without any activities like as pressing the button of smart phone. The region of pupil and iris could be fast involved with the proper iris area segmentation from the pixel value calculation of frame difference among the images which were detected with two adjacent open-eye and close-eye pictures. This proposed iris recognition could be fast processed with the proper grid size of the eye region, and designed with the frame difference between the adjacent images from the USB-OTG interface with this camera module with the restrict of searching area in face and iris location. As a result, the detection time of iris location can be reduced, and this module can be expected with eliminating the standby time of eye-open.