• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2021.06a
• /
• pp.293-296
• /
• 2021

본 논문에서는 컴퓨터 그래픽에서 주로 적용되어 왔던 푸아송 디스크 샘플링(Poisson Disk Sampling)을 3차원 영상 모델링에 적용하는 것을 제안한다. 이 샘플링 기법은 3차원 영상 센서의 핵심 기술로 사용되는 라이다 센서를 활용해 수집한 PointCloud가 특정 위치로 뭉쳐지는 클러스터 현상이 발생하지 않고 균일하게 분포하게 할 뿐 아니라 영상의 노이즈도 제거한다. Intel의 라이다 센서 L515와 Apple의 태블릿 라이다 센서를 이용해 추출한 PointCloud를 Poisson Disk Sampling 과정을 거쳐 Mesh를 생성하고 이를 SLAM 기법으로 추출한 경우와 비교한다. PointCloud의 수를 줄였을 때 더 좋은 Mesh를 생성할 수 있다.

• Proceedings of the IEEK Conference
• /
• 2002.07b
• /
• pp.1153-1156
• /
• 2002

In this parer, a high quality mesh generation method by using interval arithmetic is proposed. In the proposed method, the variance of a tangent vector at the point is considered by the automatic differentiation. From the variance, sampling points on the surface are judged whether it is adequate or not, which is calculated by the interval arithmetic. Then Delaunay triangulation is performed to the obtained sampling points, and a set of meshes is generated. The proposed method is hard to overlook the local variation of surfaces.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.35 no.4
• /
• pp.445-450
• /
• 2002

A series of 29 sampling with a 330 ${\mu}$m and a 64 ${\mu}$m mesh size of nets was conducted at a fixed station in Ilkwang Bay, southeast cost of Korea, from Oct, 2, 1991 to Oct. 10, 1992, to investigate the effects of mesh size of nets on biomass estimation of copepod Acartia steueri. The catch of copepodite and nauplius stages of A. steueli taken by two nets with different mesh size was different, showing that all developmental stages of A. steueri were retained on the 64 ${\mu}$m mesh net, but only $\geq$stage 4 copepodite were caught by the 330 ${\mu}$m mesh net. Abundance and biomass in each developmental stage estimated with the 64 ${\mu}$m mesh net were significantly higher than those of the 330 ${\mu}$m mesh net, except for adult female and stage 5 copepodite in female. The body length as well as the body width is likely to affect the catch of the nets. The mean biomass of A. steueli estimated with the traditional 330 ${\mu}$m net was 2.8 times lower than the value obtained with the 64 ${\mu}$m mesh net. However, the seasonal patterns of the biomass were comparable. These results suggest that accurate sampling strategr of the entire copepods assemblage including nauplii and copepodites are essential when estimating the abundance and biomass of copepods for the better understanding of the role of copepods in marine ecosystem.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2008.11a
• /
• pp.35-36
• /
• 2008

• Journal of Korean Society of Occupational and Environmental Hygiene
• /
• v.6 no.1
• /
• pp.97-108
• /
• 1996

A new type of passive samplers were designed and produced by authors. After evaluating the quality of activated carbon by measuring recovery rate of organic vapors and steadiness of sampling rate, activated carbon with 30 - 35 mesh produced by Company S in Korea was selected. In each passive sampler, an amount of 400 mg of the activated carbon was filled in 25-mm cassette and covered by fixed screen (or wire screen with 100 mesh). In addition to the fixed screen, a wind screen (or wire screen with 300 mesh) was also attached at outer face. The sampling rate of the new Korean passive samplers was estimated Conclusions obtained in the study are as follows. 1. Sampling rates of the newly developed Korean passive samplers were affected by sampling time. For n-hexane, sampling rates of 15- and 60-minute samples were 70.92 and 37.45 ml/min, respectively. Sampling rate of both 200- and 450-minute samples was 25.96 ml/min. It is concluded that, when passive samplers are used for measuring organic vapors, samples be collected longer than 60 minutes. 2. Sampling rate of the passive samplers was also affected by airborne concentration of organic vapors. Lower sampling rates were determined at level of 1/2 threshold limit values (TLVs) recommended by the American Conference of Governmental Industrial Hygienists (ACGIH). It is recommended that sampling rate of the passive samplers be obtained at site by measuring concentrations using both the NIOSH Method and passive samplers simultaneously. 3. When the passive samplers, which collected organic vapors, were exposed to clean air for five hours, there was no significant loss of organic vapors due to reverse diffusion. 4. When samples were stored at room temperature ($21.8{\pm}0.7^{\circ}C$) and refrigerator ($3.8{\pm}0.7^{\circ}C$), there was no significant difference in the accuracy of results. For trichloroethylene and n-hexane, accuracies were within 25 % at both temperatures until seven days. However, poor accuracy exceeding 25 % was indicated in toluene from the first day. It is recommended that samples be stored at freezing temperature below $0^{\circ}C$. 5. Sampling efficiency was significantly affected by direction of the passive samplers. Results of samplers facing wind and down, respectively, were compared. Lower amount of organic vapors were collected when the sampler was oriented down. It is recommended that, when air velocity is low in plants, the passive samplers be oriented to the wind. However, when air velocity is high, the passive samplers be oriented down.

• Journal of KIISE:Computer Systems and Theory
• /
• v.31 no.5_6
• /
• pp.297-304
• /
• 2004

In this paper, a mesh simplification algorithm based on wavelet transform and 2D planar sampling is proposed for efficient handling of 3D objects in computer applications. Since 3D vertices are directly transformed with wavelets in conventional mesh compression and simplification algorithms, it is difficult to solve tiling optimization problems which reconnect vertices into faces in the synthesis stage highly demanding vertex connectivities. However, a 3D mesh is sampled onto 2D planes and 2D polygons on the planes are independently simplified in the proposed algorithm. Accordingly, the transform of 2D polygons is very tractable and their connection information Is replaced with a sequence of vertices. The vertex sequence of the 2D polygons on each plane is analyzed with wavelets and the transformed data are simplified by removing small wavelet coefficients which are not dominant in the subjective quality of its shape. Therefore, the proposed algorithm is able to change the mesh level-of-detail simply by controlling the distance of 2D sampling planes and the selective removal of wavelet coefficients. Experimental results show that the proposed algorithm is a simple and efficient simplification technique with less external distortion.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.41 no.4
• /
• pp.279-288
• /
• 2005

The mesh selectivity of drum net traps for Buccinum opisoplectum Dall caught in the eastern sea of Korea was described. Several experiments were conducted between July 2003 and March 2004 and the data obtained from the experiment was fitted by SELECT (Share Each Length class's Catch Total) model with unequal fishing and sampling efforts. The selection curve was determined to be 'estimated fishing efficiency' by model deviance rather than AIC. However the fishing efficiency between 20mm and 35mm mesh size did not seem to be different. The $L_{50}$ lengths of the selectivity for the 55mm and 75mm mesh size were 65.4mm and 96.0mm, respectively.

• Fisheries and Aquatic Sciences
• /
• v.26 no.9
• /
• pp.548-557
• /
• 2023

The purposes of this research were to identify fish species using DNA barcodes or partial sequences of cytochrome b (Cytb) and to assess the diversity of fish in the Mae Tam reservoir, Phayao province, Thailand. Fish samples were collected 3 times, during the winter, summer, and rainy seasons, from 2 sampling sites using gillnets with 3 mesh sizes (30, 50, and 70 mm). A total of 34 representative samples were classified into 12 species, 7 families and 6 orders by morphological- and DNA barcoding-based identifications. However, one cichlid species, Cichlasoma trimaculatum, could only be identified using DNA barcoding. Family Cyprinidae had the greatest diversity, 50.00%. The diversity, richness and evenness indices ranged from 0.43-0.65, 0.64-1.46, and 0.27-0.40, respectively, indicating that fish diversity at both sampling sites was relatively low. A comparison of the catch per unit effort (CPUE) with 3 different mesh sizes found that the 50 mm mesh size was the best (474.80 ± 171.56 g/100 m²/night), followed by the 70 mm (417.41 ± 176.24 g/100 m²/night) and 30 mm mesh sizes (327.88 ± 115.60 g/100 m²/night). These results indicate that DNA barcoding is a powerful tool for species identification. Our data can be used for planning the sustainable management of fisheries resources in the Mae Tam reservoir.

• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1250-1256
• /
• 2019

Although one can easily generate real-world 3D mesh data using a 3D printer or a depth camera, the generated data inevitably includes unnecessary noise. Therefore, mesh denoising is essential to obtain intact 3D mesh data. However, conventional mathematical denoising methods require preprocessing and often eliminate some important features of the 3D mesh. To address this problem, this paper proposes a deep learning based 3D mesh denoising method. Specifically, we propose a convolution-based autoencoder model consisting of an encoder and a decoder. The convolution operation applied to the mesh data performs denoising considering the relationship between each vertex constituting the mesh data and the surrounding vertices. When the convolution is completed, a sampling operation is performed to improve the learning speed. Experimental results show that the proposed autoencoder model produces faster and higher quality denoised data than the conventional methods.

• Korean Journal of Ecology and Environment
• /
• v.53 no.2
• /
• pp.156-164
• /
• 2020

The accurate estimation of fish assemblages is highly dependent on the sampling gear used for sampling. We used data from 15 sampling sites along the Nakdong River, which is a large river in South Korea, to identify differences in assemblages and sizes of freshwater fishes collected with either cast nets or gill nets, the two most commonly used sampling gear in South Korea. The two gears differed in the fish assemblages they captured, with more species caught by gill nets. Further, due to its tighter mesh size, the cast net caught significantly smaller fishes than the gill nets(independent t-test, p<0.05). We found the cast net to be appropriate for species that inhabit shallow (less than 2 m) and open water, but inappropriate for deep water, habitats with plant beds, and nocturnal species. Thus, cast net sampling is not efficient in a large river environment, and a combination of sampling methods is more suitable for understanding fish assemblages in such habitats. In general, appropriate selection of fishing methods to specific habitats is necessary to improve data quality and minimize the misrepresentation of environmental conditions.