• International Journal of Contents
• /
• v.10 no.1
• /
• pp.23-28
• /
• 2014

When developing a classifier using various objective functions, it is important to compare the performances of the classifiers. Although there are statistical analyses of objective functions for classifiers, simulation results can provide us with direct comparison results and in this case, a comparison criterion is considerably critical. A Receiver Operating Characteristics (ROC) graph is a simulation technique for comparing classifiers and selecting a better one based on a performance. In this paper, we adopt the ROC graph to compare classifiers trained by mean-squared error, cross-entropy error, classification figure of merit, and the n-th order extension of cross-entropy error functions. After the training of feed-forward neural networks using the CEDAR database, the ROC graphs are plotted to help us identify which objective function is better.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 1998.04a
• /
• pp.187-193
• /
• 1998

In the tandom pencil slat saw lines, feeding of cedar blocks is often stopped because excessive motor current is required in a saw machine. These events are called "kickoffs' in factory lines. Kickoffs decrease productivity due to machine down-time and damage to saw blades often accompanies them. Researches on saw behavior at kickoff are required to understand and reduce the frequency and severity of kickoff events. This research aims at understanding the fundamental mechanisms of kickoffs during cutting, predicting the impending kickoff and evolving design improvements for high cutting performance with fewer and less severe kickoffs.offs.

• Journal of Forest and Environmental Science
• /
• v.31 no.2
• /
• pp.109-118
• /
• 2015

Allelopathy is an ecological phenomenon that refers to the beneficial or harmful effects of one plant on another plant, both crop and weed species, by the release of organic chemicals (allelochemicals) from plant parts by leaching, root exudation, volatilization, residue decomposition in soil and other processes in both natural and agricultural systems. Allelopathy can affect many aspects of plant ecology including occurrence, growth, plant succession, the structure of plant communities, survival, dominance, diversity, and plant productivity. In this review, we describe the concept of allelopathy, some mechanisms of operation within plants and then focus on a select number of gymnospermous tree genera: Ephedra, Pinus, Taxus, Cedrus, Juniperus, Picea, Cunninghamia and Araucaria. Pinus, Taxus (yew) and Cedrus (cedar) trees have a strong negative allelopathic effect on the germination, growth, or development of other plant species in the forest community.

• Journal of Environmental Science International
• /
• v.11 no.5
• /
• pp.445-454
• /
• 2002

Material exposure experiments were performed to evaluate the relationship between air pollution and material corrosion rates based on collaboration with researchers in China, Japan, and Korea. Qualitative and quantitative atmospheric corrosion was estimated from damage caused to bronze, copper, steel, marble, cedar, cypress, and lacquer plates exposed to outdoor and indoor conditions in certain East Asian cities. The effects of atmospheric and meteorological factors on the damage to the copper plates and marble pieces were estimated using a regression analysis. The results indicated that sulfur dioxide produced the most destruction of the materials, especially in South Korea and China. In Japan, the copper plates were damaged as a result of natural conditions and sea salt. Copper was also found to be damaged by the surface deposition of sulfur and chlorine. Meanwhile, marble was substantially degraded by gaseous sulfur dioxide, yet sulfate ions in rain had no effect. Accordingly, the analysis of air pollution from the perspective of material damage was determined to be very useful in evaluating and substantiating the intensity of air pollution in East Asia.

• Journal of the Korean Wood Science and Technology
• /
• v.33 no.5 s.133
• /
• pp.7-12
• /
• 2005

The objective of this study is to estimate the mechanical behavior in bending and the anatomical changes of wood under several deflection rates. Sample specimens of water-saturated Japanese cedar (Cryptomeria japonica) were stressed to rupture under several deflection rates. Mechanical properties of wood such as modulus of elasticity, modulus of rupture and stress at proportional limit, and anatomical changes affected by deflection rates were estimated. Microscopic observations on compression side of the test specimens when the specimen was loaded to rupture were carried out by the SEM (scanning electron microscopy). The results are summarized as follows: 1. The mechanical properties of wood were affected by variations of the deflection rates. The modulus of elasticity (MOE), modulus of rupture (MOR) and stress at proportional limit were in proportion to the logarithm of deflection rates. 2. The deflection of wood at rupture in bending increased as deflection rates decreased. 3. The variations of the microscopic deformations of sample specimens were closely related to the deflection of wood at rupture. In case of largely deflected wood by maximum bending load, severe and abundant microscopic deformations were observed.

• Journal of KSNVE
• /
• v.9 no.6
• /
• pp.1210-1217
• /
• 1999

This paper presents schematic process of identifying the principal cause to make the vibration problem of rotating circular saws. In the tandem pencil slat saw lines, feeding of cedar blocks is often stopped because excessive motro current is required in a saw motor. These events are called "kick-offs" in technical reports. Research on saw behavior at kick-offs is required to understand are reduce the frequency and severity of kick-offs events. This research aims at finding out the principal cause of kick-offs, and evloving design improvements for high cutting performance with fewer and less severe kick-offs. Measurements of critical speed, cutting force, cutting temeprature and dynamic displacements are carried out to observe the instability mechanism and also to obtain saw design parameters for the numerical analyses. And the numerical analyses involving FEM and multiple scale method are utilized to show the possibility of the principal cause.pal cause.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.9B
• /
• pp.1775-1784
• /
• 1999

This paper proposes a method of removing unnecessary hidden nodes by a new cost function that evaluates the variance and the mean of hidden node outputs during training. The proposed cost function makes necessary hidden nodes be activated and unnecessary hidden nodes be constants. We can remove the constant hidden nodes without performance degradation. Using the CEDAR handwritten digit recognition, we have shown that the proposed method can remove the number of hidden nodes up to 37.2%, with higher recognition rate and shorter learning time.

• International Journal of Computer Science & Network Security
• /
• v.21 no.7
• /
• pp.324-330
• /
• 2021

Agriculture is one of the essential needs of human life on planet Earth. It is the source of food and earnings for many individuals around the world. The economy of many countries is associated with the agriculture sector. Lots of diseases exist that attack various fruits and crops. Apple Tree Leaves also suffer different types of pathological conditions that affect their production. These pathological conditions include apple scab, cedar apple rust, or multiple diseases, etc. In this paper, an automatic detection framework based on deep learning is investigated for apple leaves disease classification. Different pre-trained models, VGG16, ResNetV2, InceptionV3, and MobileNetV2, are considered for transfer learning. A combination of parameters like learning rate, batch size, and optimizer is analyzed, and the best combination of ResNetV2 with Adam optimizer provided the best classification accuracy of 94%.

• Korean Journal of Soil Science and Fertilizer
• /
• v.26 no.4
• /
• pp.230-235
• /
• 1993

This study was carried out to investigate the supplying capacity of various forms(Water soluble, exchangeable, and available) of potassium in soils of different site quality by using surface soils from the Japanese cedar(Cryptomeria japonica) forests in Cheju island. Major results can be summerized as follows. Using continuous leaching methods with 0.01N-HCl, accumulated amounts of available potassium from surface soils of site upper and site low were about 0.6me/100g and 0.4me/100g, respectively. The release ratio of available potassium in the first $1{\ell}$ leaching of 0.01N-HCl to the total available potassium leached with $6{\ell}$ of 0.01N-HCl was over 80% for both sites, and this suggester that surface soil of both sites had the weak potassium adsorptivity as the typical volcanic ash soil. The ratio of the exchangeable potassium to the water soluble potassium was 1 : 1 for both sites. The ratio of the exchangeable potassium to the available potassium was 1 : 1 for the site low but the smaller ratio value for the site upper, indicating that the potassium supplying power of the site upper was greater than that of the site low. Available potassium was highly correlated with exchangeable Ca($0.83^{**}$) and Mg($0.84^{**}$).

• Spatial Information Research
• /
• v.14 no.4 s.39
• /
• pp.363-377
• /
• 2006

Mean annual soil loss was calculated and critical soil erosion areas were identified for the Congaree River Basin in South Carolina, USA using the Revised Universal Soil Loss Equation (RUSLE) model. In the RUSLE model, the mean annual soil loss (A) can be calculated by multiplying rainfall-runoff erosivity (R), soil erodibility (K), slope length and steepness (LS), crop-management (C), and support practice (P) factors. The critical soil erosion areas can be identified as the areas with soil loss amounts (A) greater than the soil loss tolerance (T) factor More than 10% of the total area was identified as a critical soil erosion area. Among seven subwatersheds within the Congaree River Basin, the urban areas of the Congaree Creek and the Gills Creek subwatersheds as well as the agricultural area of the Cedar Creek subwatershed appeared to be exposed to the risk of severe soil loss. As a prototype model for examining future effect of human and/or nature-induced changes on soil erosion, the RUSLE model customized for the area was embedded into ESRI ArcGIS ArcMap 9.0 using Visual Basic for Applications. Using the embedded model, users can modify C, LS, and P-factor values for each subwatershed by changing conditions such as land cover, canopy type, ground cover type, slope, type of agriculture, and agricultural practice types. The result mean annual soil loss and critical soil erosion areas can be compared to the ones with existing conditions and used for further soil loss management for the area.