• Korean Journal of Remote Sensing
• /
• v.23 no.5
• /
• pp.385-391
• /
• 2007

Subsurface Internal Waves(IWs) can be detected in satellite images as periodic alternating brighter/darker stripes. It is known that there are two types of IWs-depression type and elevation type-depending on the water depth in stratified oceans. In this study, we have quantitatively verified the process of converting polarity from depression waves to elevation waves using ERS-2 SAR image acquired over the northern South China Sea. We simulated the evolution of IWs near a turning point with a numerical model for internal wave propagation. The simulation results near the turning point clearly showed us not only a conversion process of IWs from depression to elevation waves, but also a similar wave pattern with the observed SAR image. We also simulated SAR intensity variation near the turning point. The upper layer currents were computed at regular intervals using the numerical model, as the IWs were passing through the turning point. Then, an integrated hydrodynamic-electromagnetic model was used for simulating SAR intensity profiles from the upper layer currents. The simulated SAR intensity profiles were compared with the observed SAR intensities.

• Journal of Korea Multimedia Society
• /
• v.24 no.11
• /
• pp.1552-1559
• /
• 2021

Convolution neural networks (CNNs) show notable performance in image processing and are used as representative core models. CNNs extract and learn features from large amounts of train dataset. In general, it has a structure in which a convolution layer and a fully connected layer are stacked. The core of CNN is the convolution layer. The size of the kernel used for feature extraction and the number that affect the depth of the feature map determine the amount of weight parameters of the CNN that can be learned. These parameters are the main causes of increasing the computational complexity and memory usage of the entire neural network. The most computationally expensive components in CNNs are fully connected and spatial convolution computations. In this paper, we propose a Fourier Convolution Neural Network that performs the operation of the convolution layer in the Fourier domain. We work on modifying and improving the amount of computation by applying the fast fourier transform method. Using the MNIST dataset, the performance was similar to that of the general CNN in terms of accuracy. In terms of operation speed, 7.2% faster operation speed was achieved. An average of 19% faster speed was achieved in experiments using 1024x1024 images and various sizes of kernels.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.10a
• /
• pp.514-517
• /
• 2005

Microstereolithography(MSL) has evolved from the stereolithography technique, and is also based on a light-induced layer-stacking fabrication. Although integral MSL allows the manufacture of a complete layer by one irradiation only, there is a problem related with shape precision due to the light-intensity distribution of focused image. In this study, we developed the integral MSL apparatus using Digital Micromirror Device ($DMD^{TM})$, Texas Instruments) as dynamic pattern generator. It is composed of Xenon-Mecury lamp, optical devices, pattern generator, precision stage, controllers and the control program. Also, we have studied curing depth and width of photocurable resin according to the change of exposure energy.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.3
• /
• pp.909-914
• /
• 2024

This study suggests a quantitative evaluation of transfer learning, which is widely used in various AI fields, including image recognition for robot vision. Quantitative and qualitative analyses of results applying transfer learning are presented, but transfer learning itself is not discussed. Therefore, this study proposes a quantitative evaluation of transfer learning itself based on MNIST, a handwritten digit database. For the reference network, the change in recognition accuracy according to the depth of the transfer learning frozen layer and the ratio of transfer learning data and pre-training data is tracked. It is observed that when freezing up to the first layer and the ratio of transfer learning data is more than 3%, the recognition accuracy of more than 90% can be stably maintained. The transfer learning quantitative evaluation method of this study can be used to implement transfer learning optimized according to the network structure and type of data in the future, and will expand the scope of the use of robot vision and image analysis AI in various environments.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2010.04a
• /
• pp.389-391
• /
• 2010

The Free surface influenced the wake behind a circular cylinder and its effects were investigated experimentally in a circulating water channe with the variation of water depth Instantaneous velocity fields were measured in this paper. the measured results has been compared with each other to investigate the flow characteristics of the circular cylinder's 2-dimensional section at $Re=1.0\times10^3$ using 2-frame grey level cross correlation PIV method. The flow around the circular cylinder with free surface affected the wake structure. especially, in case of d=1.0D, the boundary layer was measured in the whole area. The separation point and boundary layer of the circular cylinder could be controlled by the water depth.

• Ocean and Polar Research
• /
• v.30 no.4
• /
• pp.407-418
• /
• 2008

Seawater samples were collected at discrete depths from five stations across the polar front in the Drake Passage (Antarctic Ocean) by the $20^{th}$ Korea Antarctic Research Program in December, 2006. Nitrate concentrations of seawater increase with depth within the photic zone above the depth of Upper Circumpolar Deep Water (UCDW). In contrast, ${\delta}^{15}N$ values of seawater nitrate decrease with depth, showing a mirror image to the nitrate variation. Such a distinct vertical variation is mainly attributed to the degree of nitrate assimilation by phytoplankton as well as organic matter degradation of sinking particles within the surface layer. The preferential $^{14}{NO_3}^-$ assimilation by the phytoplankton causes $^{15}{NO_3}^-$ concentration to become high in a closedsystem surface-water environment during the primary production, whereas more $^{14}{NO_3}^-$ is added to the seawater during the degradation of sinking organic particles. The water-mass mixing seems to play an important role in the alteration of ${\delta}^{15}N$ values in the deep layer below the UCDW. Across the polar front, nitrate concentrations of surface seawater decrease and corresponding ${\delta}^{15}N$ values increase northward, which is likely due to the degree of nitrate utilization during the primary production. Based on the Rayleigh model, the calculated ${\varepsilon}$ (isotope effect of nitrate uptake) values between 4.0%o and 5.8%o were validated by the previously reported data, although the preformed ${\delta}^{15}{{NO_3}^-}_{initial}$ value of UCDW is important in the calculation of ${\varepsilon}$ values.

• Journal of Navigation and Port Research
• /
• v.34 no.5
• /
• pp.331-336
• /
• 2010

The free surface influenced the wake behind a circular cylinder and its effects were investigated experimentally in a circulating water channel with the variation of water depth. Instantaneous velocity fields were measured in this paper. The measured results has been compared with each other to investigate the flow characteristics of the circular cylinder's 2-dimensional section at $Re=1.0{\times}10^3$ using 2-frame grey level cross correlation PIV method. The flow around the circular cylinder with free surface affected the wake structure. Especially, in case of d=1.0D, the boundary layer was measured in the whole area. The separation point and boundary layer of the circular cylinder could be controlled by the water depth.

• Electrical & Electronic Materials
• /
• v.9 no.5
• /
• pp.463-469
• /
• 1996

The double insulating layer consisting of anodic oxide and ZnS was formed for HgCdTe metal insulator semiconductor(MIS) structure. ZnS was evaporated on the anodic oxide grown in H$_{2}$O$_{2}$ electrolyte. Recently, this insulating mechanism for HgCdTe MIS has been deeply studied for improving HgCdTe surface passivation. It was found through TEM observation that an interface layer is formed between ZnS and anodic oxide layers for the first time in the study of this area. EDS analysis of chemical compositions using by electron beam of 20.angs. in diameter and XPS depth composition profile indicated strongly that the new interface is composed of ZnO. Also TEM high resolution image showed that the structure of oxide layer has been changed from the amorphous state to the microsrystalline structure of 100.angs. in diameter after the evaporation of ZnS. The double insulating layer with the resistivity of 10$^{10}$ .ohm.cm was estimated to be proper insulating layer of HgCdTe MIS device. The optical reflectance of about 7% in the region of 5.mu.m showed anti-reflection effect of the insulating layer. The measured C-V curve showed the large shoft of flat band voltage due to the high density of fixed oxide charges about 1.2*10$^{12}$ /cm$^{2}$. The oxygen vacancies and possible cationic state of Zn in the anodic oxide layer are estimated to cause this high density of fixed oxide charges.

• Proceedings of the Korean Society of Medical Physics Conference
• /
• 2003.09a
• /
• pp.44-44
• /
• 2003

As a basic measurement tool in the areas of animal models of human disease, gene expression and therapy, and drug discovery and development, small animal PET imaging is being used increasingly. An ideal small animal PET should have high sensitivity and high and uniform resolution across the field of view to achieve high image quality. However, the combination of long narrow pixellated crystal array and small ring diameter of small animal PET leads to the degradation of spatial resolution for the source located at off center. This degradation of resolution can be improved by determining the depth of interaction (DOI) in the crystal and by taking into account the information in sorting the coincident events. Among a number of 001 identification schemes, dual layer phsowich detector has been widely investigated by many research groups due to its practicability and effectiveness on extracting DOI information. However, the effects of each crystal length composing dual layer phoswich detector on DOI measurements and image qualities were not fully characterized. In order to minimize the DOI effect, the length of each layer of phoswich detector should be optimized. The aim of this study was to perform simulations using a simulation tool, GATE to design the optimum lengths of crystals composing a dual layer phoswich detector. The simulated small PET system employed LSO front layer LuYAP back layer phoswich detector modules and the module consisted of 8${\times}$8 arrays of dual layer crystals with 2 mm ${\times}$ 2 mm sensitive area coupled to a Hamamatsu R7600 00 M64 PSPMT. Sensitivities and variation of radial resolutions were simulated by varying the length of LSO front layer from 0 to 10 mm while the total length (LSO + LuYAP) was fixed to 20 mm for 10 cm diameter ring scanner. The radial resolution uniformity was markedly improved by using DOI information. There existed the optimal lengths of crystal layers to minimize the variation of radial resolutions. In 10 cm ring scanner configuration, the radial resolution was kept below 3.4 mm over 8 cm FOV while the sensitivity was higher than 7.4% for LSO 5 mm : LuYAP 15 mm phoswich detector. In this study, the optimal length of dual layer phoswich detector was derived to achieve high and uniform radial resolution.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.1
• /
• pp.21-29
• /
• 2013

The current software for creating and editing animation offers the possibility of dividing variable objects into layers. This paper suggests a method to render and edit each of right and left layer in the process of creating a 3D stereoscopic animation. It is possible to apply a variety of effects to left and right images and to adjust the gap of binocular parallax through this method so that wider variation of images and richer sense of depth can be produced. In addition, this method offers freedom of deleting and inserting an object, controlling size and position of the object and adjusting the gap of binocular parallax. Thus, new sense of depth can be obtained. This method shows the results to apply various effects on the 3D stereoscopic image and to reduce visual fatigue rather than the method to render simultaneously both of left and right layer.