• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.7
• /
• pp.1380-1386
• /
• 2007

This paper proposed an automatic recognition system of shipping container identifiers using fuzzy-based noise removal method and ART2-based self-organizing supervised learning algorithm. Generally, identifiers of a shipping container have a feature that the color of characters is blacker white. Considering such a feature, in a container image, all areas excepting areas with black or white colors are regarded as noises, and areas of identifiers and noises are discriminated by using a fuzzy-based noise detection method. Areas of identifiers are extracted by applying the edge detection by Sobel masking operation and the vertical and horizontal block extraction in turn to the noise-removed image. Extracted areas are binarized by using the iteration binarization algorithm, and individual identifiers are extracted by applying 8-directional contour tacking method. This paper proposed an ART2-based self-organizing supervised learning algorithm for the identifier recognition, which improves the performance of learning by applying generalized delta learning and Delta-bar-Delta algorithm. Experiments using real images of shipping containers showed that the proposed identifier extraction method and the ART2-based self-organizing supervised learning algorithm are more improved compared with the methods previously proposed.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.4
• /
• pp.772-780
• /
• 2008

K-means is one of the simplest unsupervised learning algorithms that solve the clustering problem. However K-means suffers the basic shortcoming: the number of clusters k has to be known in advance. In this paper, we propose extensions of X-means, which can estimate the number of clusters using Bayesian information criterion(BIC). We introduce two different versions of algorithm: modified X-means(MX-means) and generalized X-means(GX-means), which employ one full covariance matrix for one cluster and so can estimate the number of clusters efficiently without severe over-fitting which X-means suffers due to its spherical cluster assumption. The algorithms start with one cluster and try to split a cluster iteratively to maximize the BIC score. The former uses K-means algorithm to find a set of optimal clusters with current k, which makes it simple and fast. However it generates wrongly estimated centers when the clusters are overlapped. The latter uses EM algorithm to estimate the parameters and generates more stable clusters even when the clusters are overlapped. Experiments with synthetic data show that the purposed methods can provide a robust estimate of the number of clusters and cluster parameters compared to other existing top-down algorithms.

• Journal of agriculture & life science
• /
• v.44 no.6
• /
• pp.171-182
• /
• 2010

Water demand and resource at a watershed scale were investigated to prepare for long-term water planning of the Namgang sub-basin. The quantity of water resource was defined as average annual runoff by a simple Tank model with three serial tanks, and water demand for public, industrial, agricultural and the other uses was determined using the per-unit method employed in the Water Vision 2020, the Ministry of Construction and Transportation. The results showed that total amount of water resources in the Namgang sub-basin was estimated as about $935,414{\times}10^{3}m^3/yr$ for a 10-year period from 2000 to 2009 and the water withdrawals in public, industrial, agricultural and other sector were derived as $105,493{\times}10^{3}m^3/yr$, $32,686{\times}10^{3}m^3/yr$, $243,194{\times}10^{3}m^3/yr$, and $81,615{\times}10^{3}\;m^3m^3/yr$, respectively. In addition, the Namgaram Inno-city project could increase the overall water demand by $17,156{\times}10^{3}\;mm^3/yr$ due to the population influx.

• Korean Journal of Remote Sensing
• /
• v.21 no.2
• /
• pp.113-120
• /
• 2005

Oceanographic measurements from Ieodo Ocean Research Station and its vicinity were compared for assessment and mutually adjusted with satellite data. From the Topex/Poseidon and ERS-1/2 radar altimeter and scatterometer data, sea surface height, wind speed and direction were extracted and analyzed. Shipborne wind direction data acquired in June 1995 show good coherence with the satellite data, while sea surface height and wind speed show differences, possibly resulting from the distance between the measurement points. This can be improved by analyzing more satellite data or using other available shipborne data. The recent 3 months of Ieodo Station data between December 2004 and February 2005 were also analyzed and compared with the satellite data. The Ieodo Station data were found to have considerable gaps during the period as well as seriously biased particular when the data were averaged with some abnormal data. The Ieodo Station and satellite data were then mutually adjusted on the basis of their statistics. Ieodo Station oceanographic measurements are very efficient for ground-frothing of satellite data because they are stationary and the station is located far from the coast. On the other hand, the satellite measurements are the only data to fill up gaps and adjust biases of the Ieodo Station data.

• Journal of the Korean Recycled Construction Resources Institute
• /
• v.7 no.3
• /
• pp.224-234
• /
• 2019

Waterproof layers are installed in civil engineering structures and bridge construction is commonly finished by applying a layer of regular or asphalt concrete above the waterproof layer. However, asphalt materials are susceptible to melting at high temperature due to its superior temperature sensitivity, and this causes the waterproofing material to melt due to the high temperature of the asphalt concrete, thereby increasing the defect occurrence rate due to the thickness reduction. In this study, tensile strength and elongation of hard and soft type of MMA(Methyl Methacrylate) applied to bridges were compared in accordance to standard performance criteria based on different mixture ratios. Results of comparative testing showed that hard MMA resin can display a satisfactory tensile strength, and soft MMA resin displays satisfactory elongation properties, but as the two resin types are separately used, neither types are able to satisfy the standard requirements outlined in KS F 4932. When the amount of the powder exceeds 56.25% of the total amount, voids are generated on the surface after curing and self leveling was impossible and a heterogeneous surface is formed. Furthermore, when the hard resin: soft resin: powder mixture ratio was set to 15g: 85g: 150g. the tensile strength was $1.5N/mm^2$ and the elongation percentage was 133% which satisfy the tensile performance of KS F 4932.

• Geophysics and Geophysical Exploration
• /
• v.1 no.1
• /
• pp.8-18
• /
• 1998

In exploration seismology, the Kirchhoff hyperbola has been successfully used to migrate reflection seismo-grams. The mathematical basis of Kirchhoff hyperbola has not been clearly defined and understood for the application of prestack or poststack migration. The travel time from the scatterer in the subsurface to the receivers (exploding reflector model) on the surface can be a kinematic approximation of Green's function when the source is excited at position of the scatterer. If we add the travel time from the source to the scatterer in the subsurface to the travel time of exploding reflector model, we can view this travel time as a kinematic approximation of the partial derivative wavefield with respect to the velocity or the density in the subsurface. The summation of reflection seismogram along the Kirchhoff hyperbola can be evaluated as an inner product between the partial derivative wavefield and the field reflection seismogram. In addition to this kinematic interpretation of Kirchhoff hyperbola, when we extend this concept to shallow refraction seismic data, the stacking of refraction data along the straight line can be interpreted as a measurement of an inner product between the first arrival waveform of the partial derivative wavefield and the field refraction data. We evaluated the Kirchhoff hyperbola and the straight line for stacking the refraction data in terms of the first arrival waveform of the partial derivative wavefield with respect to the velocity or the density in the subsurface. This evaluation provides a firm and solid basis for the conventional Kirchhoff migration and the straight line stacking of the refraction data.

• Elastomers and Composites
• /
• v.38 no.3
• /
• pp.227-234
• /
• 2003

Silicone composites for high voltage insulator (HVI SC) were prepared by adding aluminum trihydrate(ATH) treated by surface treatment agent to base silicone compound at the ratio oi 100:20, 100:40, 100:60, 100:80, and 100:100, respectively And also, ATH was treated by various surface treatment agents, such as stearic arid, acryl silane, and vinyl silane under compounding process. Mechanical properties and electrical properties were investigated for the various contents of ATH and surface-treatment agents. Mechanical properties such as tensile strength, elongation, and tear strength decreased as the load of ATH increased. Volume resistivity, AC break down strength, and tracking resistance for HVI SC containing ATH treated by vinyl silane were better than those for HVI SC containing ATH treated by other surface treatment agents, such as stearic acid and acryl silane. Polymer-filler interaction of silicone composites according to surface treatment agents was studied by measuring bound rubber contend(BR). From the experimental results, BR of silicone composite containing ATH treated with vinyl silane was higher than that of the others. The degree of rule for silicone composite was investigated using Rheometer. Maximum torque of silicone composites contaning ATH treated with vinyl silane was higher than that of silicone composite contaning ATH treated with other surface agents.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.25 no.5
• /
• pp.149-156
• /
• 2021

FRP has been proposed to replace the steel as a reinforcement in the concrete structures for addressing the corrosion issue. However, FRP-Rebar does not behave in the same manner as steel because the properties of FRP are different. For example, FRP-Rebar has a high tensile strength, low stiffness, and linear elastic behavior which results in a difference bonding mechanism to transfer the load between the reinforcement and the surrounding concrete. Therefore, bonding behavior between FRP-Rebar and concrete has to be investigated using the bonding test. So, Pull-out test has been used to estimate bond behavior because it is simple. However, the results of pull-out test have a difference with flexural-boding behavior of FRP-Rebar concrete member. So, it is needed to evaluate the real fleuxral-bonding behavior. In this study, the evaluation method to flexural-bonding behavior of FRP-Rebar concrete member was reviewed and compared. It was found that the most accurate evaluation method for the fleuxral-bonding behavior of FRP-Rebar concrete member was splice beam test, however, the size and length of specimen have to increase than other test method and the design and analysis of splice beam is complex. Meanwhile, characteristics of concrete could be reflected by using arched beam test, unlike hinged beam test which is based on the ignored change of moment arm length in equilibrium equation. However, the possibility of shear failure exists before the flexural-bonding failure occur.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.2
• /
• pp.15-29
• /
• 2021

The signal intersection is the most challenging space for autonomous vehicles. To promote the safe driving of autonomous vehicles on urban roads with traffic signals, autonomous vehicles need to receive traffic signal information from infrastructure through V2I communication. Thus, a protocol for providing traffic signal information was added to the standard traffic signal controller specification of the National Police Agency. On the other hand, there are technical limitations when applying this to digital traffic signal controllers because the protocols are defined mainly for analog traffic signal controllers. Therefore, this study proposes developing a signal information linkage module to provide traffic signal information from a digital traffic signal controller to an autonomous vehicle and an algorithm improvement method that can provide accurate traffic signal information at the time of traffic signal transition.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.18 no.2
• /
• pp.58-76
• /
• 2019

As the number of reported injuries has tended to increase over time, large hospitalization expenditure from excessive medical treatments and hospitalization, and insurance frauds associated with moral hazard in minor collisions have caused a global societal problem. Many occupants of rear-ended vehicles involved in rear-end collisions complain of whiplash injury, which is also known as neck injury, without any anatomical and radiological evidence. With only clinical symptoms, stating that a whiplash injury is a type of injury defined by the Abbreviated Injury Scale would be difficult. Therefore, this study focuses on minor rear-end collisions, where the rear-ender vehicle collides with the rear-ended vehicle at rest. The mathematics dynamic model is employed to simulate a total of 100 rear-end collision scenarios based on various weights and collision speeds and identify how the weights and speeds of both vehicles influence the risk of whiplash injury in occupants involved in minor rear-end collisions. The possibility of an injury is very high when the same-weight vehicles are involved in accidents at collision speeds of 15 km/h or higher. The possibilities are 36% and 84% with collision speeds of 15 km/h and 20 km/h, respectively, if weights are disregarded.