• Journal of KIISE:Software and Applications
• /
• v.37 no.12
• /
• pp.942-945
• /
• 2010

In order to reduce the efforts to inspect the reported alarms from a static buffer overflow analyzer, we present an effective method to filter out redundant alarms. In the static analysis, a sequence of multiple alarms are frequently found due to the same cause in the code. In such a case, it is sufficient and reasonable for programmers to examine the first alarm instead of the entire alarms in the same sequence. Based on this observation, we devise a buffer overflow analysis that filters out redundant alarms with our context refinement technique. Our experiment with several open source programs shows that our method reduces the reported alarms by 23% on average.

• Journal of KIISE:Software and Applications
• /
• v.41 no.7
• /
• pp.492-500
• /
• 2014

This paper presents the performance enhancement of automatic bilingual lexicon extraction by using refinement of pivot-context vectors under the standard pivot-based approach, which is very effective method for less-resource language pairs. In this paper, we gradually improve the performance through two different refinements of pivot-context vectors: One is to filter out unhelpful elements of the pivot-context vectors and to revise the values of the vectors through bidirectional translation probabilities estimated by Anymalign and another one is to remove non-noun elements from the original vectors. In this paper, experiments have been conducted on two different language pairs that are bi-directional Korean-Spanish and Korean-French, respectively. The experimental results have demonstrated that our method for high-frequency words shows at least 48.5% at the top 1 and up to 88.5% at the top 20 and for the low-frequency words at least 43.3% at the top 1 and up to 48.9% at the top 20.

• ETRI Journal
• /
• v.45 no.5
• /
• pp.822-835
• /
• 2023

Recent semantic segmentation frameworks usually combine low-level and high-level context information to achieve improved performance. In addition, postlevel context information is also considered. In this study, we present a Context ReFinement Network (CRFNet) and its training method to improve the semantic predictions of segmentation models of the encoder-decoder structure. Our study is based on postprocessing, which directly considers the relationship between spatially neighboring pixels of a label map, such as Markov and conditional random fields. CRFNet comprises two modules: a refiner and a combiner that, respectively, refine the context information from the output features of the conventional semantic segmentation network model and combine the refined features with the intermediate features from the decoding process of the segmentation model to produce the final output. To train CRFNet to refine the semantic predictions more accurately, we proposed a sequential training scheme. Using various backbone networks (ENet, ERFNet, and HyperSeg), we extensively evaluated our model on three large-scale, real-world datasets to demonstrate the effectiveness of our approach.

• Journal of KIISE:Computer Systems and Theory
• /
• v.30 no.11
• /
• pp.629-654
• /
• 2003

A new multiresolutional scheme that reconstructs a polygonal mesh from the set of contours is presented. In the first step, we apply a radial gradient method to extract the contours on the sampled slices from a volume data. After classifying the types of the edges on the contours, we represent the contour using the context-free grammar. The polygons between two neighboring contours are generated through the traversal of the derivation trees of the context-free grammar. The polygonal surface of the coarsest resolution is refined through the refinement of the contours, which is executed by casting more rays on the slices. The topologies between the polygonal surfaces of various resolutions are maintained from the fact that the radial gradient method preserves the topologies of the contours of various resolutions. The proposed scheme provides efficient computation and compression methods for the tiling procedure with the feature preservation.

• Journal of The Korean Association For Science Education
• /
• v.23 no.3
• /
• pp.276-285
• /
• 2003

In the earlier study, Park (2002) described the process of student' conceptual change as a successive refinement and articulation of their conceptual framework. In this study, the process of conceptual change for three university students were analyzed more in depth. As results, six types of conceptual change through successive refinement and articulation were observed: (1) original conception was elaborated in detail, (2) conception was differentiated according to the context, (3) some conceptions were re-explained theoretically after construction it based on experimental data, (4) non-coherent conceptions in the early stage get to have coherency in the later stage, (5) model of explanation gets to be complicated by excluding ideal conditions, (6) qualitative explanations were changed into quantitative ones.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.4
• /
• pp.938-958
• /
• 2024

Unsupervised Video Object Segmentation (UVOS) is a highly challenging problem in computer vision as the annotation of the target object in the testing video is unknown at all. The main difficulty is to effectively handle the complicated and changeable motion state of the target object and the confusion of similar background objects in video sequence. In this paper, we propose a novel deep Dual-stream Co-enhanced Network (DC-Net) for UVOS via bidirectional motion cues refinement and multi-level feature aggregation, which can fully take advantage of motion cues and effectively integrate different level features to produce high-quality segmentation mask. DC-Net is a dual-stream architecture where the two streams are co-enhanced by each other. One is a motion stream with a Motion-cues Refine Module (MRM), which learns from bidirectional optical flow images and produces fine-grained and complete distinctive motion saliency map, and the other is an appearance stream with a Multi-level Feature Aggregation Module (MFAM) and a Context Attention Module (CAM) which are designed to integrate the different level features effectively. Specifically, the motion saliency map obtained by the motion stream is fused with each stage of the decoder in the appearance stream to improve the segmentation, and in turn the segmentation loss in the appearance stream feeds back into the motion stream to enhance the motion refinement. Experimental results on three datasets (Davis2016, VideoSD, SegTrack-v2) demonstrate that DC-Net has achieved comparable results with some state-of-the-art methods.

• Journal of Daesoon Thought and the Religions of East Asia
• /
• v.2 no.1
• /
• pp.133-155
• /
• 2022

Throughout the Chinese sphere, that is, in the People's Republic of China and Taiwan, tea houses, tea stores, and tea meditation groups often without a particular religious orientation have appeared in the past decades. Tea lovers groups with various philosophical discussions appear throughout cyberspace, where people show their appreciation for tea as a drinkable delicious product and a spiritual tool. The question to be asked here is whether it should be recognized as a religious or spiritual practice agent in and of itself? Should we then talk about the present-day movement of teaism? If we do recognize this as a spiritual phenomenon, should it then be labeled as a New Religious Movement? The trajectory of tea in China is intrinsically connected to religious traditions. This connection is historical, yet it plays a part in the contemporary religious and spiritual sphere. The article explores the continuation and developments of tea culture in the context of the religious sphere of China, looking at practices connected to tea of communities, religious organizations, and individuals. The author explores how tea drinking, commercializing and tea related practices intersect with politics, materiality, and spirituality in contemporary society. In this context it is then argued that tea is a cultural element, religious self-refinement tool, and an active material agent with social-political capacities. The study includes historical narratives, ethnographic data, and literary sources about tea, making up a genealogy of tea which encompasses ritualistic aspects, economic aspects, and power relations related to tea in Chinese society.

• The Bulletin of The Korean Astronomical Society
• /
• v.38 no.2
• /
• pp.101.2-101.2
• /
• 2013

We present a new computational fluid dynamic (CFD) simulation code. The code employs the moving and polyhedral unstructured mesh scheme, which is known as a superior approach to the conventional SPH (smoothed particle hydrodynamics) and AMR (adaptive mesh refinement) schemes. The code first generates unstructured meshes by the Voronoi tessellation at every time step, and then solves the Riemann problem for surfaces of every Voronoi cell to update the hydrodynamic states as well as to move former generated meshes. For the second-order accuracy, the MUSCL-Hancock scheme is implemented. To increase efficiency for generating Voronoi tessellation we also develop the incremental expanding method, by which the CPU time is turned out to be just proportional to the number of particles, i.e., O(N). We will discuss the applications of our code in the context of cosmological simulations as well as numerical experiments for galaxy formation.

• Journal of Korean Institute of Industrial Engineers
• /
• v.20 no.1
• /
• pp.99-112
• /
• 1994

This paper describe the development of a systematic method of slicing solid parts based on a data structure called Sliced Profile Data Structure(SPDS). SPDS is an augmented polygon data structure that allows multiple layers of sliced profiles to be connected together. The method consists of five steps: (1) Selection of slicing directions, (2) Determination of slicing levels, (3) Creation of sliced profiles, (4) Connection of sliced profiles, and (5) Refinement. The presented method is aimed at enhancing the applicability of CSG for manufacturing by overcoming the problem of non-uniqueness and global nature. The SPDS-based method of feature extraction is suitable for recognizing broad scope of features with detailed information. The method is also suitable for identifying the global relationships among features and is capable of incorporating the context dependency of feature extraction.

• The Bulletin of The Korean Astronomical Society
• /
• v.39 no.2
• /
• pp.65.2-65.2
• /
• 2014

We present a new hydrodynamic simulation code based on the Voronoi tessellation for estimating the density precisely. The code employs both of Lagrangian and Eulerian description by adopting the movable mesh scheme, which is superior to the conventional SPH (smoothed particle hydrodynamics) and AMR (adaptive mesh refinement) schemes. The code first generates unstructured meshes by the Voronoi tessellation at every time step, and then solves the Riemann problem for all surfaces of each Voronoi cell so as to update the hydrodynamic states as well as to move current meshes. Besides, the IEM (incremental expanding method) is devised to compute the Voronoi tessellation to desired degree of speed, thereby the CPU time is turned out to be just proportional to the number of particles, i.e., O(N). We discuss the applications of our code in the context of cosmological simulations as well as numerical experiments for galaxy formation.