• Smart Media Journal
• /
• v.9 no.1
• /
• pp.23-29
• /
• 2020

This paper presents an approach for dynamic hand gesture recognition by using algorithm based on 3D Convolutional Neural Network (3D_CNN), which is later extended to 3D Residual Networks (3D_ResNet), and the neural network based key frame selection. Typically, 3D deep neural network is used to classify gestures from the input of image frames, randomly sampled from a video data. In this work, to improve the classification performance, we employ key frames which represent the overall video, as the input of the classification network. The key frames are extracted by SegNet instead of conventional clustering algorithms for video summarization (VSUMM) which require heavy computation. By using a deep neural network, key frame selection can be performed in a real-time system. Experiments are conducted using 3D convolutional kernels such as 3D_CNN, Inflated 3D_CNN (I3D) and 3D_ResNet for gesture classification. Our algorithm achieved up to 97.8% of classification accuracy on the Cambridge gesture dataset. The experimental results show that the proposed approach is efficient and outperforms existing methods.

• The Journal of the Korea Contents Association
• /
• v.8 no.5
• /
• pp.94-103
• /
• 2008

In this paper, we adapt the $\kappa$-means clustering technique to segmenting a given 3D mesh. In order to avoid the locally minimal convergence and speed up the computing time, first we extract sharp vertices from the mesh by analysing its curvature and convexity that respectively reflect the local and global geometric characteristics from the viewpoint of cognitive science. Next the sharp vertices are partitioned into $\kappa$ clusters by iterated converging with the $\kappa$-means clustering method based on the geodesic distance instead of the Euclidean distance between each pair of the sharp vertices. For obtaining the effective result of $\kappa$-means clustering method, it is crucial to assign an initial value to $\kappa$ appropriately. Hence, we automatically compute a reasonable number of clusters as an initial value of $\kappa$. Finally the mesh segmentation is completed by merging other vertices except the sharp vertices into the nearest cluster by geodesic distance.

• Animal cells and systems
• /
• v.5 no.3
• /
• pp.253-262
• /
• 2001

Ligand-receptor clustering event is the most important step in leukocyte adhesion and spreading on endothelial cells. Intercellular adhesion molecule-1 (ICAM-1) has been shown to enhance leukocyte adhesion, but the molecular event during the process of adhesion is unclear. To visualize the dynamics of ICAM-1 movement during adhesion, we have engineered stable Chinese hamster ovary cell lines expressing ICAM-1 fused to a green fluorescent protein (IC1_GFP/CHO) and examined them under the fluorescence microscopy. The transfection of IC1_GFP alone in these cells was sufficient to support the adhesion of K562 cells that express $\alpha$L$\beta$2 (LFA-1) integrin stimulated by CBR LFA-1/2 mAb. This phenomenon was mediated by ICAM-1-LFA-1 interactions, as an mAb that specifically inhibits ICAM-1-LFA-1 interaction (RRl/l) completely abolished the adhesion of LFA-1＊ cells to IC1_ GFP/CHO cells. We found that the characteristic of adhesion was followed almost immediately (~10 min) by the rapid accumulation of ICAM-1 on CHO cells at a tight interface between the two cells. Interestingly, ICI_GFP/CHO cells projected plasma membrane and encircled approximately half surface of LFA-1+ cells, as defined by confocal microscopy. This unusual phenomenon was also confirmed on HUVEC after adhesion of LFA-1＊ cells. Neither cytochalasin D nor 2,3-butanedione 2-monoxime an inhibitor of myosin light chain kinase blocked LFA-1-mediated ICAM-1 clustering, indicating that actin cytoskeleton and myosin-dependent contractility are not necessary for ICAM-1 clustering. Taken together, we suggest that leukocyte adhesion to endothelial cells induces specialized form of ICAM-1 clustering that is distinct from immunological synapse mediated by T cell interaction with antigen presenting cells.

• Korean Journal of Remote Sensing
• /
• v.37 no.5_3
• /
• pp.1373-1387
• /
• 2021

Forest fire poses a significant threat to the environment and society, affecting carbon cycle and surface energy balance, and resulting in socioeconomic losses. Widely used multi-spectral satellite image-based approaches for burned area detection have a problem in that they do not work under cloudy conditions. Therefore, in this study, Sentinel-1 Synthetic Aperture Radar (SAR) data from Europe Space Agency, which can be collected in all weather conditions, were used to identify forest fire damaged area based on a series of processes including Principal Component Analysis (PCA) and K-means clustering. Four forest fire cases, which occurred in Gangneung·Donghae and Goseong·Sokcho in Gangwon-do of South Korea and two areas in North Korea on April 4, 2019, were examined. The estimated burned areas were evaluated using fire reference data provided by the National Institute of Forest Science (NIFOS) for two forest fire cases in South Korea, and differenced normalized burn ratio (dNBR) for all four cases. The average accuracy using the NIFOS reference data was 86% for the Gangneung·Donghae and Goseong·Sokcho fires. Evaluation using dNBR showed an average accuracy of 84% for all four forest fire cases. It was also confirmed that the stronger the burned intensity, the higher detection the accuracy, and vice versa. Given the advantage of SAR remote sensing, the proposed statistical processing and K-means clustering-based approach can be used to quickly identify forest fire damaged area across the Korean Peninsula, where a cloud cover rate is high and small-scale forest fires frequently occur.

• 한국전산유체공학회:학술대회논문집
• /
• 2000.10a
• /
• pp.86-91
• /
• 2000

Automatic method for generation of mesh and three dimension natural convection flow result adapted by this method are presented in this paper. It lake long time to meshing com plex 3-D geometries, and It's difficult to clustering grid at surface boundary. Octree structure resolve this difficulty.

• Journal of IKEEE
• /
• v.16 no.2
• /
• pp.95-101
• /
• 2012

A novel approach for the reconstruction of 3D building model from aerial image data is proposed in this paper. In this approach, a Centroid Neural Network (CNN) with a metric of line segments is proposed for connecting low-level linear structures. After the straight lines are extracted from an edge image using the CNN, rectangular boundaries are then found by using an edge-based grouping approach. In order to avoid producing unrealistic building models from grouping lined segments, a hierarchical grouping method is proposed in this paper. The proposed hierarchical grouping method is evaluated with a set of aerial image data in the experiment. The results show that the proposed method can be successfully applied for the reconstruction of 3D building model from satellite images.

• The Journal of the Korea Contents Association
• /
• v.10 no.12
• /
• pp.77-83
• /
• 2010

Recently, IT-based environment is changing rapidly as changing in web services platform, evolution of cloud computing environments and expanding the base of a smart market. Accordingly, monitoring development of environment is changing quickly. So a customizable SaaS-based monitoring tool is required to provide monitoring services. It has to support a variety of environmental monitoring and a resource managers with requested information, and by an enhanced monitoring framework in clouding environment of management system. In this paper, the 3D viewer for the management of sensor node management system was designed and built. Through the 3D viewer by enhancing the accessibility and visibility, the sensor network will allow resources to be used efficiently.

• Journal of Information Processing Systems
• /
• v.7 no.3
• /
• pp.413-424
• /
• 2011

A variety of different metrics has been introduced to measure the similarity of two given sequences. These widely used metrics are ranging from spell correctors and categorizers to new sequence mining applications. Different metrics consider different aspects of sequences, but the essence of any sequence is extracted from the ordering of its elements. In this paper, we propose a novel sequence similarity measure that is based on all ordered pairs of one sequence and where a Hasse diagram is built in the other sequence. In contrast with existing approaches, the idea behind the proposed sequence similarity metric is to extract all ordering features to capture sequence properties. We designed a clustering problem to evaluate our sequence similarity metric. Experimental results showed the superiority of our proposed sequence similarity metric in maximizing the purity of clustering compared to metrics such as d2, Smith-Waterman, Levenshtein, and Needleman-Wunsch. The limitation of those methods originates from some neglected sequence features, which are considered in our proposed sequence similarity metric.

• Journal of Crop Science and Biotechnology
• /
• v.11 no.1
• /
• pp.63-68
• /
• 2008

The induced genetic divergence was estimated in 44 mutant lines of finger millet variety GPU 26, developed by single and combination treatments with gamma rays, EMS and NG using three multivariate analyses. The mutant lines were grouped into eight genetically diverse clusters by multivariate D2 and canonical analyses and 11 clusters by dendrogram grouping through Gower's similarity coefficient. The clustering pattern in these three methods was almost similar. Twelve mutant lines in D2 and 13 in the dendrogram grouping method were grouped in the parental cluster(Cluster I) indicating that they did not possess enough divergence from the parent to be classified as micromutant lines. However a large proportion of mutant lines showed divergence from the parent variety and also among themselves. No definite relationship of mutagenic origin and clustering of mutant lines were observed. The mutant lines developed from the same mutagenic treatments often grouped into different clusters indicating that each mutagenic treatment was effective in inducing diverse types of changes in the nine traits studied. The hybridization program between the divergent mutant lines GE 2-2 or GE 3-4 with GG 3-1 is expected to give promising and desirable segregants in subsequent generations. Traits such as days to 50% flowering and days to maturity had major contributions to the induced genetic divergence.

• Journal of IKEEE
• /
• v.15 no.3
• /
• pp.191-197
• /
• 2011

An image segmentation model using fuzzy c-means with intuition (FCM-I) model is proposed for the segmentation of magnetic resonance image in this paper. In FCM-I, a measurement called intuition level is adopted so that the intuition level helps to alleviate the effect of noises. A practical magnetic resonance image data set is used for image segmentation experiment and the performance is compared with those of some conventional algorithms. Results show that the segmentation method based on FCM-I compares favorably to several conventional clustering algorithms. Since FCM-I produces cluster prototypes less sensitive to noises and to the selection of involved parameters than the other algorithms, FCM-I is a good candidate for image segmentation problems.