• 전자통신동향분석
• /
• 제32권6호
• /
• pp.8-16
• /
• 2017

In this paper, we review studies attempting to triumph over the limitation of Si-based semiconductor technologies through a heterogeneous integration of high mobility compound semiconductors on a Si substrate, and the co-integration of electronic and/or optical devices. Many studies have been conducted on the heterogeneous integration of various materials to overcome the Si semiconductor performance and obtain multi-purpose functional devices. On the other hand, many research groups have invented device fusion technologies of electrical and optical devices on a Si substrate. They have co-integrated Si-based CMOS and InGaAs-based optical devices, and Ge-based electrical and optical devices. In addition, chip and wafer bonding techniques through TSV and TOV have been introduced for the co-integration of electrical and optical devices. Such intensive studies will continue to overcome the device-scaling limitation and short-channel effects of a MOS transistor that Si devices have faced using a heterogeneous integration of Si and a high mobility compound semiconductor on the same chip and/or wafer.

• Journal of Information Processing Systems
• /
• 제18권6호
• /
• pp.794-802
• /
• 2022

The recovery of reasonable depth information from different scenes is a popular topic in the field of computer vision. For generating depth maps with better details, we present an efficacious monocular depth prediction framework with coordinate attention and feature fusion. Specifically, the proposed framework contains attention, multi-scale and feature fusion modules. The attention module improves features based on coordinate attention to enhance the predicted effect, whereas the multi-scale module integrates useful low- and high-level contextual features with higher resolution. Moreover, we developed a feature fusion module to combine the heterogeneous features to generate high-quality depth outputs. We also designed a hybrid loss function that measures prediction errors from the perspective of depth and scale-invariant gradients, which contribute to preserving rich details. We conducted the experiments on public RGBD datasets, and the evaluation results show that the proposed scheme can considerably enhance the accuracy of depth prediction, achieving 0.051 for log10 and 0.992 for δ<1.25³ on the NYUv2 dataset.

• 한국진공학회:학술대회논문집
• /
• 한국진공학회 2011년도 제41회 하계 정기 학술대회 초록집
• /
• pp.199-199
• /
• 2011

The addition of a carbanion to ${\yen}{\acute{a}}{\yen}{\hat{a}}$-unsaturated carbonyl compounds is of importance in the C-C bond formation reactions for modern pharmaceuticals and organic synthesis. Recently, heterogeneous asymmetric catalysis became more attractive area of research because of the easy recovery and separation of the catalyst from the reaction system. Most of synthetic methods for heterogeneous catalysts were grafting or immobilization of homogeneous catalyst onto the solid supports. Trans-1,2-Diaminocyclohexane(DACH) and L-proline ligands have been enormously used as chiral ligands in several catalytic transformation under homogenous conditions. Our group prepared l-proline functionalized mesoporous silica was synthesized under acidic condition using a poly(ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide) triblock copolymer template (EO20PO70EO20, Pluronic P-123, BASF). Furthermore, we successfully directly synthesized trans-1,2 diaminocyclohexane functionalized mesoporous silica by using microwave method. The direct functionalization of chiral ligand into the framework of mesoporous materials is expected to be useful for the heterogeneous asymmetric catalysis. So, we adopt the direct synthesis of chiral ligand functionalized mesoporous silica by using thermal and microwave irradiation. Then, chiral ligand functionalized mesoporous silicas were applied to enantioselective asymmetric catalytic reactions.

• 대한산업공학회지
• /
• 제43권1호
• /
• pp.39-48
• /
• 2017

In a digital economy, users actively express their needs in many ways. Thus, many researchers analyze what users need and whether they are satisfied or not through opinion mining. In addition, they begin to find technology opportunities in heterogeneous technology fields. But they did not connect users' opinion to technology development process, only focused on natural language processing or marketing or manufacturing area. Also, heterogeneous technology fields are focused on fusion technology. Thus, this study suggests a novel approach that is based on sentimental value and can be applied to exploring technology opportunities in heterogeneous fields. Sentimental value is calculated from users' opinion through sLDA. The heterogeneous technology opportunity is explored by patent analysis. This research contributes to suggesting a hybrid methodology through patent and users' opinion. In addition, it can provide managerial efficiency by suggesting base data onto decision making.

• Bulletin of the Korean Chemical Society
• /
• 제27권9호
• /
• pp.1381-1385
• /
• 2006

Cobalt (cyclam) complex has been successfully immobilized onto SBA-15, and proven to be an active catalyst for the epoxidation of styrene with tert-butyl hydroperoxide as a terminal oxidant. The selectivity for styrene oxide was observed to be up to 66% with 40% styrene conversion after 12h reaction time. The reversible redox cycle between Co(III) and Co(II) couple which was supposed to play key role during the epoxidation reaction was supported by a cyclic voltametry analysis. The textural properties of the catalyst was characterized by XRD, N2 adsorption-desorption, and TEM analysis.

• 대한임베디드공학회논문지
• /
• 제11권2호
• /
• pp.57-65
• /
• 2016

In this paper, we propose a system for efficient use of shared memory between CPU and GPU. The system, called Fusion Architecture, assures consistency of the shared memory and minimizes cache misses that frequently occurs on Heterogeneous System Architecture or Unified Virtual Memory based systems. It also maximizes the performance for memory intensive jobs by efficient allocation of GPU cores. To test between architectures on various scenarios, we introduce the Fusion Architecture Analyzer, which compares OpenMP, OpenCL, CUDA, and the proposed architecture in terms of memory overhead and process time. As a result, Proposed fusion architectures show that the Fusion Architecture runs benchmarks 55% faster and reduces memory overheads by 220% in average.

• 한국정보처리학회논문지
• /
• 제6권10호
• /
• pp.2728-2739
• /
• 1999

This paper proposes an effective fusion algorithm for retrieval results from heterogeneous information sources in federated digital libraries. The algorithm determines the population of documents retrieved from involved information sources for a given query and evaluates the degree of relevance between the query and the population. The evaluated results are used as relevance distribution information for collection fusion. The main informations used for the fusion are relevance distribution among collections, the population size N, and ranking information of relevant documents in their origin. We also present th performance evaluation of the algorithm by developing the prototype of a meta-searcher.

• 한국의상디자인학회지
• /
• 제9권2호
• /
• pp.167-178
• /
• 2007

The purpose of this study is to systematically examine fusion phenomenon in the cultural aspect aiming to grasp the characteristics of fusion phenomenon, which has a steady impact on fashion trend. A reason for fusion to be noted as a cultural code is because diversely cultural interpretations are impossible from one viewpoint. Therefore, it is aimed to inquire into the characteristics of fusion culture, which is one phenomenon of newly mass culture, and to examine which cultural fusion phenomenon in modern fashion is being displayed in fashion design. First, the culturally fusion phenomenon, which was indicated in modern fashion, is displayed as repetition. The fusion, which integrates into fashion as a cultural code, is a trend of pursuing the experience with different culture. The needs of consumers are shown as fusion fashion that repeatedly uses the traditional detail and the modern detail through cultural repetition. Second, it is displayed as transformation. A concept of fusion is not the simple aggregate in [1+1=2], but the creation of new culture called [1+1=1]. In modern fashion, the cultures in the East and the West are harmoniously transformed, and through transformation, each culture is being compressively delivered. Third, it is displayed as interdependency. The fusion fashion which was newly re-interpreted while having reciprocal influence upon mutual culture, is re-interpreted by forming the equal relationship in heterogeneous elements without destroying fashion that was made previously.

• Mycobiology
• /
• 제29권1호
• /
• pp.15-18
• /
• 2001

Protoplast fusion is a useful technique for establishing fungal hybrids to overcome the natural barriers. The ultrastructure of protoplast and its fusion process were observed using a scanning electron microscopy(SEM) and a transmission electron microscopy(TEM). The protoplasts were variable in size from $0.5{\sim}15{\mu}m$ in diameter, and the mean diameter was about $3{\sim}5{\mu}m$. It was impossible to discriminate protoplasts of Lentinula edodes from protoplasts of Coriolus versicolor by size and surface structure. Big aggregates of the dehydrated protoplasts were observed, after polyethylene glycol 4000 treatment. Nucleus, mitochondria, lipid granules and various vesicles having granules were scattered in the cytoplasm. The vesicles were heterogeneous in size and vary from one protoplast to another. The fused membrane layer of the two protoplasts was observed. Time protoplast membrane contact and reorganization of membrane components were essential condition for protoplast fusion. Transmission electron micrograph showed fused protoplasts and flattening of the cells in the area of the membrane contact. We hope that our electron microscopic observations provide some insights into the understanding of the fusion process of protoplast in fungi.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권4호
• /
• pp.819-833
• /
• 2013

Computing global features based on local features using a bio-inspired framework has shown promising performance. However, for some tough applications with large intra-class variances, a single local feature is inadequate to represent all the attributes of the images. To integrate the complementary abilities of multiple local features, in this paper we have extended the efficacy of the bio-inspired framework, HMAX, to adapt heterogeneous features for global feature extraction. Given multiple global features, we propose an approach, designated as parameterized metric learning, for high dimensional feature fusion. The fusion parameters are solved by maximizing the canonical correlation with respect to the parameters. Experimental results show that our method achieves significant improvements over the benchmark bio-inspired framework, HMAX, and other related methods on the Caltech dataset, under varying numbers of training samples and feature elements.