• The Journal of the Convergence on Culture Technology
• /
• v.8 no.1
• /
• pp.545-550
• /
• 2022

In Deep Learning method, it is well known that it requires large amount of data to train the deep neural network. And it also requires the labeling of each data to fully train the neural network, which means that experts should spend lots of time to provide the labeling. To alleviate the problem of time-consuming labeling process, some methods have been suggested such as weak-supervised method, one-shot learning, self-supervised, suggestive learning, and so on. In this manuscript, those methods are analyzed and its possible future direction of the research is suggested.

• Journal of Korean Neurosurgical Society
• /
• v.66 no.4
• /
• pp.476-481
• /
• 2023

Objective : The latest version of the Leksell Gamma Knife Icon^TM allows for mask- and frame-based fixation. Although mask fixation provides fractionated treatment and immobilization using a noninvasive method, it is not free from collision. The authors investigated the collision problem with a modified mask fixation method. Methods : This study presents a case of two meningiomas in the frontal area, where a collision occurs in the occipital area. A modified mask fixation method was introduced to avoid the collision : first, the edges of the head cushion were cut off and polystyrene beads with a diameter of approximately 5 cm were removed. Next, the head cushion was sealed using a stapler. Finally, the head cushion was flattened in the adapter. We compared the shot coordinates, 3-dimensional (3D) error, clearance distance, and vertical depth of the head cushion between the initial and modified mask fixations. Results : When comparing the initial and modified mask fixations, the difference in the shot coordinates was +10.5 mm along the y-axis, the difference in the 3D error was approximately 18 mm, and the difference in clearance was -10.2 mm. The head cushion was approximately 8 mm deeper in the modified mask fixation. Conclusion : Based on these findings, we recommend a modified mask fixation method for gamma knife radiosurgery using ICON with a collision.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1998.06b
• /
• pp.92-97
• /
• 1998

For the efficient searching and browsing of digital video, it is essential to extract the internal structure of the video contents. As an example, a news video consists of several sections such as politics, economics, sports and others, and also each section consists of individual topics. With this information in hand, users can ore easily access the required video frames. This paper addresses the problem of automatic shot boundary detection and selection of representative frames (R-frames), which are the essential step in recognizing the internal structure of video contents. In the shot boundary detection, a new algorithm that have dual detectors which are designed specifically for the abrupt boundaries (cuts) and gradually changing bounaries respectively is proposed. Compared to the existing 미algorithms that mostly have tried to detect both types by a single mechanism, the proposed algorithm is proved to be more robust and accurate. Also in the problem of R-frame selection, simple mechanical approaches such as selecting one frame every other second have been adopted. However this approach often selects too many R-frames in static short, while drops important frames in dynamic shots. To improve the selection mechanism, a new R-frame selection algorithm that uses motion information extracted from pixel difference is proposed.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.2
• /
• pp.366-372
• /
• 2020

The Army's rocket missile is a one-shot weapon system, which is produced and used for only one mission, and requires high reliability. While reliability analysis with failure data can result in underestimation of the life distribution, reliability analysis with all the non-failure data can result in overestimation of the life distribution. Under or overestimation of the life distribution can lead to cost increase by early disposal or complete observation of all rocket missiles. In order to overcome this problem, the Army suggests the guideline of the number of samples from non-failure data for reliability analysis with failure data. However, the currently used sampling method can generate errors for predicting the failure rate. To solve this problem, this study proposes a new sampling procedure for predicting a future failure rate using non-failure data. The comparison test between the currently used sampling method and the proposed sampling method is conducted and the result shows that the proposed sampling method can predict the future failure rate more accurately.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.7
• /
• pp.271-278
• /
• 2021

As the current cyber attacks become more intelligent, the existing Intrusion Detection System is difficult for detecting intelligent attacks that deviate from the existing stored patterns. In an attempt to solve this, a model of a deep learning-based intrusion detection system that analyzes the pattern of intelligent attacks through data learning has emerged. Intrusion detection systems are divided into host-based and network-based depending on the installation location. Unlike network-based intrusion detection systems, host-based intrusion detection systems have the disadvantage of having to observe the inside and outside of the system as a whole. However, it has the advantage of being able to detect intrusions that cannot be detected by a network-based intrusion detection system. Therefore, in this study, we conducted a study on a host-based intrusion detection system. In order to evaluate and improve the performance of the host-based intrusion detection system model, we used the host-based Leipzig Intrusion Detection-Data Set (LID-DS) published in 2018. In the performance evaluation of the model using that data set, in order to confirm the similarity of each data and reconstructed to identify whether it is normal data or abnormal data, 1D vector data is converted to 3D image data. Also, the deep learning model has the drawback of having to re-learn every time a new cyber attack method is seen. In other words, it is not efficient because it takes a long time to learn a large amount of data. To solve this problem, this paper proposes the Siamese Convolutional Neural Network (Siamese-CNN) to use the Few-Shot Learning method that shows excellent performance by learning the little amount of data. Siamese-CNN determines whether the attacks are of the same type by the similarity score of each sample of cyber attacks converted into images. The accuracy was calculated using Few-Shot Learning technique, and the performance of Vanilla Convolutional Neural Network (Vanilla-CNN) and Siamese-CNN was compared to confirm the performance of Siamese-CNN. As a result of measuring Accuracy, Precision, Recall and F1-Score index, it was confirmed that the recall of the Siamese-CNN model proposed in this study was increased by about 6% from the Vanilla-CNN model.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.6
• /
• pp.133-140
• /
• 2005

This paper presents a rapid face shape acquisition system. The system is composed of two cameras and one projector. The technique works by projecting a pattern on the object and capturing two images with two cameras. We use a 'one shot' system which provides 3D data acquired by single image per camera. The system is good for rapid data acquisition as our purpose. We use the 'absolutely coded pattern' using the hue and saturation of pattern lines. In this 'absolutely coded pattern' all patterns have absolute identification numbers. We solve the correspondence problem between the two images by using epipolar geometry and absolute identification numbers. In comparison to the 'relatively coded pattern' which uses relative identification numbers, the 'absolutely coded pattern' helps obtain rapid 3D data by one to one point matching on an epipolar line. Because we use two cameras, we obtain two images which have similar hue and saturation. This enables us to have the same absolute identification numbers in both images, and we can use the absolutely coded pattern for solving the correspondence problem. The proposed technique is applied to face data and the total time for shape acquisition is estimated.

• Journal of Korean Ophthalmic Optics Society
• /
• v.13 no.1
• /
• pp.65-69
• /
• 2008

Purpose: To develop a dot sight device for a sighting shot using the doublet reflector. Methods: We designed the singlet reflector and the doublet reflector by Sigma 2000 program and compared the one to the other. Results: In analysis of finite ray aberration, we could confirm that it has the effective field of view with the free-parallax which is 3.3 times wider than the existing dot sight device using the singlet reflector. If you use the doublet reflector, the central thickness of optical system become more than two times bigger than the existing one. As a result, when the image of a target object is made on the observer's retina, if you make the first side's radius of curvature equal with the second one as the case of the singlet reflector, changes in magnification are appeared. To conquer this problem, we had to make be satisfied with the afocal condition in the case of doublet reflector. Conclusions: we could develop the dot sight device for a sighting shot using the doublet reflector which has the effective field of view with the free-parallax which is 3.3 times wider than the existing dot sight device using the singlet reflector.

• The Korean Journal of Pesticide Science
• /
• v.12 no.1
• /
• pp.1-8
• /
• 2008

The widespread and diverse sulfonylurea (SU) resistance problem has found in Korea, where one-shot-treatment herbicides such as pylazosulfuron/molinate and bensulfuron/molinate have been used continuously since 1989. The SU-resistant weeds of 7 annual weeds and 3 perennial weeds as of 2008 have confirmed in paddy fields in Korea. An effective management to SU-resistant weeds requires an integrated approach toward the weed control system, in particular, as to the drastic changes of herbicides development. Recent trend of new paddy herbicides in Japan has been developing to maximize the management of SU-resistant weeds. In the future, it is expected that the development of paddy herbicides in Korea is likely to be shifted toward the new "one-shot-treatment" included with herbicides of over 3-ways to maximize the control of resistant weeds. Bromobutide and carfentrazone are effective against sedges and broad-leaved weeds, respectively, and benzobicyclone and pyrimisulfam are effective against sedges and broad-leaved weeds.

• Bulletin of the Korean Chemical Society
• /
• v.31 no.8
• /
• pp.2293-2298
• /
• 2010

Focused electrospray (FES) deposition method is presented for matrix-assisted laser desorption/ionization (MALDI) mass spectrometry. FES ion optics consists of two cylindrical focusing electrodes capped with a truncated conical electrode through which an electrospray emitter passes along the cylindrical axis. A spray of charged droplets is focused onto a sample well on a MALDI target plate under atmospheric pressure. The shape and size distributions of matrix crystals are visualized by scanning electron microscope and the mass spectra are obtained by time-of-flight mass spectrometry. Angiotensin II, bradykinin, and substance P are used as test samples, while $\alpha$-cyano-4-hydroxycinnamic acid and dihydroxybenzoic acid are employed as matrices. FES of a sample/matrix mixture produces fine crystal grains on a 1-3 mm spot and reproducibly yields the mass spectra with little shot-to-shot and spot-to-spot variations. Although FES greatly stabilizes the signals, the space charge due to matrix ions limits the detection sensitivity of peptides. To avoid the space charge problem, we adopted a dual FES/FES mode, which separately deposits matrix and sample by FES in sequence. The dual FES/FES mode reaches the detection sensitivity of 0.88 amol, enabling ultrasensitive of peptides by homogeneously depositing matrix and sample under atmospheric pressure.

• Korean Chemical Engineering Research
• /
• v.48 no.4
• /
• pp.490-498
• /
• 2010

The optimum mold design and the optimum process condition were constructed upon executing process simulation of rubber injection molding with the commercial CAE program of MOLDFLOW(Ver. 5.2) in order to solve the process-problems of K company relating to air-traps and short-shots. The former occurs at the cavity edge of torque-rod-bush and the latter takes place for the injection molding of dynamic dampers. As a result the process problem relating to air traps was solved by optimizing edge-angle and the number of gates to prevent the flow congestion of flow-front and to make the flow-front movement unaffected by congestion. For dynamic dampers of K company the unmolded flaw caused by their unfilled cavity was corrected by installing the air-vent at the confronting locations of both upstream and downstream of flow-front where air traps frequently occur. Besides the unmolded flaws were rectified by altering the position of gate from the upper to the middle or by increasing the number of gates. Thus the process problems of K company relating to air-traps and short-shots of torque-rod-bush and dynamic dampers, respectively, were solved by proper altering of mold design with process simulation of rubber injection molding.