• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.36 no.5
• /
• pp.335-341
• /
• 2018

The KDE (Kernel Density Estimation) technique in GIS (Geographic Information System) has been widely used as a method for determining whether a phenomenon occurring in space forms clusters. Most human-generated events such as traffic accidents and retail stores are distributed according to a road network. Even if events on forward and rear roads have short Euclidean distances, network distances may increase and the correlation between them may be low. Therefore, the NKDE (Network-based KDE) technique has been proposed and applied to the urban space where a road network has been developed. KDE is being studied in the field of business GIS, but there is a limit to the microscopic analysis of economic activity along a road. In this study, the NKDE technique is applied to the analysis of urban phenomena such as the density of shops rather than traffic accidents that occur on roads. The results of the NKDE technique are also compared to pedestrian networks and road centerline networks. The results show that applying NKDE to microscopic trade area analysis can yield relatively accurate results. In addition, it was found that pedestrian network data that can consider the movement of actual pedestrians are necessary for accurate trade area analysis using NKDE.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.32 no.3
• /
• pp.173-181
• /
• 2019

To estimate probabilistic distribution function from experimental data, kernel density estimation(KDE) is mostly used in cases when data is insufficient. The estimated distribution using KDE depends on bandwidth selectors that smoothen or overfit a kernel estimator to experimental data. In this study, various bandwidth selectors such as the Silverman's rule of thumb, rule using adaptive estimates, and oversmoothing rule, were compared for accuracy and conservativeness. For this, statistical simulations were carried out using assumed true models including unimodal and multimodal distributions, and, accuracies and conservativeness of estimating distribution functions were compared according to various data. In addition, it was verified how the estimated distributions using KDE with different bandwidth selectors affect reliability analysis results through simple reliability examples.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2011.05a
• /
• pp.808-810
• /
• 2011

The background estimation algorithms had a significant impact on the performance of image processing and recognition. In this paper, background estimation algorithms were analysis of complexity and performance as preprocessing of image recognition. It was evaluated the performance of Gaussian Running Average, Mixture of Gaussian, and KDE algorithm. The simulation results show that KDE algorithm outperforms compared to the other algorithms.

• Korean Journal of Environment and Ecology
• /
• v.28 no.6
• /
• pp.642-649
• /
• 2014

Mallard (Anas platyrhynchos) is the abundant winter visitor in South Korea. Mallard migrates long distances between Russian Siberia and Korea. This species prefers a rice paddy area as their winter habitat. We captured birds using cannon-net, and attached the GPS-Mobile phone based Telemetry(WT-200) on Seven Mallards in the winter of 2011~2013. We were monitored wintering home-range and movement distance. We analyzed the tracking location data using ArcGIS 9.0 and calculated Kernel Density Estimation (KDE) and Minimum Convex Polygon (MCP). The average home-range in the wintering ground by MCP was $118.8km^2$(SD=70.1, n=7)and the maximum home-rang was $221.8km^2$ and the minimum was $27.7km^2$. Extents of home-range by KDE were $60.0km^2$(KDE 90%), $23.0km^2$(KDE 70%) and $11.6km^2$(KDE 50%). Mallard moved an average of 19.4 km from start site(attach to WT-200 site), maximum moved was 33.2 km and minimum moved was 9.4 km. The average distance of 0.8 km between GPS fixed point(range 0.2~1.6 km), maximum moved was 19.7 km. Mallard moved a very short distance in wintering season and showed a very high water-dependent trends in wintering site.

• The Korean Journal of Applied Statistics
• /
• v.24 no.6
• /
• pp.987-994
• /
• 2011

From the point view of credit evaluation whose population is divided into the default and non-default state, two methods are considered to estimate conditional distribution functions: one is to estimate under the assumption that the data is followed the mixture normal distribution and the other is to use the kernel density estimation. The parameters of normal mixture are estimated using the EM algorithm. For the kernel density estimation, five kinds of well known kernel functions and four kinds of the bandwidths are explored. In addition, the corresponding ROC functions are obtained based on the estimated distribution functions. The goodness-of-fit of the estimated distribution functions are discussed and the performance of the ROC functions are compared. In this work, it is found that the kernel distribution functions shows better fit, and the ROC function obtained under the assumption of normal mixture shows better performance.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.8
• /
• pp.1659-1664
• /
• 2011

Fuzzy C-Means (FCM) is one of the most widely used clustering algorithms and has been used in many applications successfully. However, FCM has some shortcomings and initial prototype selection is one of them. As FCM is only guaranteed to converge on a local optimum, different initial prototype results in different clustering. Therefore, much care should be given to the selection of initial prototype. In this paper, a new initialization method for FCM using kernel density estimation (KDE) is proposed to resolve the initialization problem. KDE can be used to estimate non-parametric data distribution and is useful in estimating local density. After KDE, in the proposed method, one initial point is placed at the most dense region and the density of that region is reduced. By iterating the process, initial prototype can be obtained. The initial prototype such obtained showed better result than the randomly selected one commonly used in FCM, which was demonstrated by experimental results.

• Korean Journal of Environment and Ecology
• /
• v.27 no.1
• /
• pp.11-21
• /
• 2013

We tried to find out the core feeding site of the Red-crowned Crane(Grus japonensis) in Cheorwon, Korea by using analysis techniques which are MCP(minimum convex polygon), KDE(kernel density estimation), LoCoH(local nearest-neighbor convex-hull). And, We discussed the difference and meaning of result among analysis methods. We choose the data of utilization distribution from distribution map of Red-crowned Crane in Cheorwon, Korea at $17^{th}$ February 2012. Extent of the distribution area was $140km^2$ by MCP analysis. Extents of core feeding area of the Red-crowned Crane were $33.3km^2$($KDE_{1000m}$), $25.7km^2$($KDE_{CVh}$), $19.7km^2$($KDE_{LSCVh}$), according to the 1000m, CVh, LSCVh in value of bandwidth. Extent, number and shape complexity of the core area has decreased, and size of each core area have decreased as small as the bandwidth size(default:1000m, CVh: 554.6m, LSCVh: 329.9). We would suggest the CVh value in KDE analysis as a proper bandwidth value for the Red-crowned crane's core area zoning. Extent of the distribution range and core area have increased and merged into the large core area as a increasing of k value in LoCoH analysis. Proper value for the selecting core area of Red-crowned Crane's distribution was k=24, and extent of the core area was $18.2km^2$, 16.5% area of total distribution area. Finally, the result of LoCoH analysis, we selected two core area, and number of selected core area was smaller than selected area of KDE analysis. Exact value of bandwidth have not been used in studies using KDE analysis in most articles and presentations of the Korea. As a result, it is needed to clarify the exact using bandwidth value in KDE studies.

• Korean Journal of Environment and Ecology
• /
• v.33 no.1
• /
• pp.1-8
• /
• 2019

Pintail (Anas acuta) is the major wintering bird in South Korea and known as a major mediator of the highly pathogenic avian influenza (HPAI). Pintail migrates long distances between Russian Siberia and Korea. This species prefers a rice paddy area as their winter habitat. The purpose of this study is to provide the data necessary for the conservation and management of bird habitats in Korea by understanding the wintering home-range and habitat of pintail in Korea. We captured six pintails using a cannon-net in the winter of 2015 and attached the GPS-mobile phone based telemetry (WT-300) on them to study the wintering home-range and wintering habitat. We analyzed the tracking location data using ArcGIS 9.0 Animal Movement Extension and calculated Kernel Density Estimation (KDE) and Minimum Convex Polygon (MCP). The average home-range in the wintering ground analyzed by MCP was $677.3km^2$ (SD=130.2, n=6) while the maximum and minimum were $847.7km^2$ and $467.5km^2$, respectively. Extents of home-range analyzed by KDE were $194.7km^2$ (KDE 90%), $77.4km^2$ (KDE 70%), and $35.3km^2$ (KDE 50%). The pintails mostly used both sea and paddy field as habitat in the winter season and utilized paddy fields more during the nighttime and than the daytime. We concluded that the home-range and habitat of pintails in the winter could be used as the reference data for the preservation of species, management of habitats, and coping with a breakout of HPAI.

• Korean Journal of Environment and Ecology
• /
• v.30 no.2
• /
• pp.165-172
• /
• 2016

Mallard and Spot-billed ducks that are typical wintering water birds use the wide rice field in the center of the water system as a wintering ground. I try to figure out the daily movement distance from Home-Range to wintering ground for mallard and spot-billed ducks in Mangyunggang, Dongjingang where located in central region. In 2015 wintering period by using a Cannon-net, I attached WT-300 to 5 mallard and 5 spot-billed ducks. Daily movement distance is an overall average 0.89km, the largest distance was 31.09km. Daily movement distance of mallard was 0.97km, the largest distance was 28.78km. Daily movement distance of Spot-billed ducks was 0.80km, the largest distance was 33.39km. Home-Range analysis is used by the SHP files that is compatible with GIS and ArcGIS 9.0 Animal Movement Extension, it was analyzed using the Minimum Convex Polygon Method (MCP) and the Kernel Density Estimation (KDE). The behavior rights of two kinds of ducks was $490.34km^2$ by Minimum Convex Polygon Method (MCP) ($SD=311.20km^2N=10$), an important habitats Kernel Density Estimation (KDE 50%) was $42.24km^2$. Home-Range of Mallard (MCP) was $568.02km^2$, it is wider than home-range(MCP) of spot-billed duck $397.13km^2$ relatively, the core habitats of mallard is $53.05km^2$, it is wider than mallard's core habitats(KDE 50%) $29.26km^2$ relatively.

• Proceedings of the National Institute of Ecology of the Republic of Korea
• /
• v.2 no.4
• /
• pp.285-292
• /
• 2021

We investigated habitat use and home range of a rescued and released white-naped crane using GPS tracking technology in Cheorwon, South Korea, from October 2016 to March 2017. Four types of roosting sites were identified: frozen reservoirs, paddy fields, rivers, and wetlands. Upon arrival, the white-naped crane preferred wetlands in the Demilitarized Zone (DMZ). In late wintering season, it showed a tendency to change main roosting sites in the following order: rice paddies, rivers, and frozen reservoirs. Among 14 sleeping places, Civilian Control Zone (CCZ) with various type of available habitats was more preferred than the DMZ. Places outside of CCZ were rarely used due to anthropogenic disturbances during the night. The tracked white-naped crane widely chose daytime feeding sites while moving around all over rice paddies in the CCZ. Mean diurnal movement distance was 10.5 km with a maximum of 24.8 km. Its home range measured with Minimum Convex Polygon (MCP) and Kernel Density Estimation (KDE) was 172.30 km² with MCP, 159.60 km² with KDE 95%, 132.48 km² with KDE 90%, and 42.45 km² with KDE 50%. All estimated values of home ranges were higher in the early and later winter than those in the middle period.