• Journal of the Korean Society of Industry Convergence
• /
• v.23 no.6_2
• /
• pp.1025-1032
• /
• 2020

In this paper, we propose a method to classify road weather conditions into rain, fog, and sun using a SVM (Support Vector Machine) classifier after extracting weather features from images acquired in real time using an optical sensor installed on a roadside post. A multi-dimensional weather feature vector consisting of factors such as image sharpeness, image entropy, Michelson contrast, MSCN (Mean Subtraction and Contrast Normalization), dark channel prior, image colorfulness, and local binary pattern as global features of weather-related images was extracted from road images, and then a road weather classifier was created by performing machine learning on 700 sun images, 2,000 rain images, and 1,000 fog images. Finally, the classification performance was tested for 140 sun images, 510 rain images, and 240 fog images. Overall classification performance is assessed to be applicable in real road services and can be enhanced further with optimization along with year-round data collection and training.

• Structural Engineering and Mechanics
• /
• v.83 no.1
• /
• pp.79-92
• /
• 2022

In this paper, the impact of a vernacular pozzolanic kaolinite mine on concrete alkali-silica reaction and strength has been evaluated. For making the samples, kaolinite powder with various levels has been used in the quality specification test of aggregates based on the ASTM C1260 standard in order to investigate the effect of kaolinite particles on reducing the reaction of the mortar bars. The compressive strength, X-Ray Diffraction (XRD) and Scanning Electron Microscope (SEM) experiments have been performed on concrete specimens. The obtained results show that addition of kaolinite powder to concrete will cause a pozzolanic reaction and decrease the permeability of concrete samples comparing to the reference concrete specimen. Further, various machine learning methods have been used to predict ASR-induced expansion per different amounts of kaolinite. In the process of modeling methods, optimal method is considered to have the lowest mean square error (MSE) simultaneous to having the highest correlation coefficient (R). Therefore, to evaluate the efficiency of the proposed model, the results of the support vector machine (SVM) method were compared with the decision tree method, regression analysis and neural network algorithm. The results of comparison of forecasting tools showed that support vector machines have outperformed the results of other methods. Therefore, the support vector machine method can be mentioned as an effective approach to predict ASR-induced expansion.

• The Journal of Korean Society for Radiation Therapy
• /
• v.20 no.1
• /
• pp.45-53
• /
• 2008

Purpose: We are trying to analyze 130 patients' conditions by using our Helical Tomotherapy, which was installed in our center in Oct. 2007. We will be statistically approach this examination and analyze so that we will be able to figure out adaptive plans according to the change in place of the tumor, GTV (gross tumor volume), total amount of time it took, vector (${\upsilon}=\surd$x2+y2+z2) and the change in size of the tumor. Materials and Methods: Objectives were the patients who were medicated with Tomotherapy in our medical center since Oct. 2007 August 2008. The Average age of the patients were 53 years old (Minimum 25 years old, Maximum 83 years old). The parts of the body we operated were could be categorized as Head&neck (n=22), Chest (n=47), Abdomen (n=25), Pelvis (n=11), Bone (n=25). MVCT had acted on 2702 times, and also had acted on our adaptive plan toward patients who showed big difference in the size of tumor. Also, after equalizing our gained MVCT and kv-CT we checked up on the range of possible mistake, using x, y, z, roll and vector. We've also investigated on Set-up, MVCT, average time of operation and target volume. Results: Mean time on table was 22.8 minutes. Mean treatment time was 13.26 minutes. Mean correction (mm) was X=-0.7, Y=-1.4, Z=5.77, roll=0.29, vector=8.66 Head&neck patients had 2.96 mm less vector value in movement than patients of Chest, Abdomen, Bone. In increasing order, Head&neck, Bone, Abdomen, Chest, Pelvis showed the vector value in movement. Also, there were 27 patients for adaptive plan, 39 patients, who had long or multiple tumor. We could know that When medical treatment is one cure plan, it takes 32 minutes, and when medical treatment is two cure plan, it takes 40 minutes that one medical treatment takes 21 minutes, and the other medical treatment takes 19 minutes. Conclusion:With our basic tools, we could bring more accurate IMRT with MVCT. Also, through our daily image, we checked up on the change in tumor so that adaptive plan could work. It was made it possible to take the cure of long or multiple tumor, the cure in a nearby OAR, and the complicated cure that should make changes of gradient dose distribution.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 1999.10a
• /
• pp.313-320
• /
• 1999

In this paper, a translation, rotation and scale invariant system for the classification of closed 2D images using the bispectrum of a contour sequence and the weighted fuzzy mean method is derived and compared with the classification process using one of the competitive neural algorithm, called a LVQ(Learning Vector Quantization). The bispectrun based on third order cumulants is applied to the contour sequences of the images to extract fifteen feature vectors for each planar image. These bispectral feature vectors, which are invariant to shape translation, rotation and scale transformation, can be used to represent two-dimensional planar images and are fed into an classifier using weighted fuzzy mean method. The experimental processes with eight different shapes of aircraft images are presented to illustrate the high performance of the proposed classifier.

• The KSFM Journal of Fluid Machinery
• /
• v.5 no.4 s.17
• /
• pp.19-25
• /
• 2002

The flow characteristics in a bifurcated duct are investigated experimentally. Physical properties such as mean velocity vectors, mean vorticity, and total pressure distributions are obtained for three different Reynolds numbers (578, 620, 688) using PIV measurements and CFD analysis. Also, two different dividing ducts ($90^{\circ},\;60^{\circ}$) were selected for study. The results of this study would be useful to the engineers designing flow systems for heating, ventilation, air conditioning and waste-water purification plants.

• Journal of the Korea Computer Industry Society
• /
• v.5 no.5
• /
• pp.755-760
• /
• 2004

In this paper, an advanced successive elimination algorithm was proposed using mean absolute difference of neighboring search points. By using mean absolute difference of neighboring search points, the search point in motion estimation can be eliminated effeciently without matching evaluation that requires very intensive computations. By using adaptive MAD calculation algorithm, the candidate matching block can be eliminated early. So, the number of the proposed algrorithm was verified by experimental results.

• Bulletin of the Korean Mathematical Society
• /
• v.54 no.1
• /
• pp.159-175
• /
• 2017

We solve the $Bj{\ddot{o}}rling$ problem for constant mean curvature one surfaces in hyperbolic three-space and in de Sitter three-space. That is, we show that for any regular, analytic (and spacelike in the case of de Sitter three-space) curve ${\gamma}$ and an analytic (timelike in the case of de Sitter three-space) unit vector field N along and orthogonal to ${\gamma}$, there exists a unique (spacelike in the case of de Sitter three-space) surface of constant mean curvature 1 which contains ${\gamma}$ and the unit normal of which on ${\gamma}$ is N. Some of the consequences are the planar reflection principles, and a classification of rotationally invariant CMC 1 surfaces.

• Information Systems Review
• /
• v.9 no.1
• /
• pp.85-103
• /
• 2007

The purpose of this study is to evaluate the performance of collaborative filtering recommender algorithms for better prediction accuracy of the customer's preference. The accuracy of customer's preference prediction is compared through the MAE of neighborhood based collaborative filtering algorithm and correspondence mean algorithm. It is analyzed by using MovieLens 1 Million dataset in order to experiment with the prediction accuracy of the algorithms. For similarity, weight used in both algorithms, commonly, Pearson's correlation coefficient and vector similarity which are used generally were utilized, and as a result of analysis, we show that the accuracy of the customer's preference prediction of correspondence mean algorithm is superior. Pearson's correlation coefficient and vector similarity used in two algorithms are calculated using the preference rating of two customers' co-rated movies, and it shows that similarity weight is overestimated, where the number of co-rated movies is small. Therefore, it is intended to increase the accuracy of customer's preference prediction through expanding the number of the existing co-rated movies.

• Journal of Korea Multimedia Society
• /
• v.6 no.5
• /
• pp.788-796
• /
• 2003

This paper proposes using dynamic momentum for squential learning method. Using The dynamic momentum improves convergence speed and performance by the variable momentum, also can identify it in the RMSE(root mean squared error). The proposed method is reflected using variable momentum according to current state. While static momentum is equally influenced on the whole, dynamic momentum algorithm can control the convergence rate and performance. According to the variable change of momentum by training. Unlike former classification and regression problems, this paper confirms both performance and regression rate of the dynamic momentum. Using RMSE(root mean square error ), which is one of the regression methods. The proposed dynamic momentum has been applied to the kernel adatron and kernel relaxation as the new sequential learning method of support vector machine presented recently. In order to show the efficiency of the proposed algorithm, SONAR data, the neural network classifier standard evaluation data, are used. The simulation result using the dynamic momentum has a better convergence rate, performance and RMSE than those using the static moment, respectively.

• Journal of Korean Society for Atmospheric Environment
• /
• v.22 no.5
• /
• pp.614-626
• /
• 2006

Using the data of three environmental monitoring sites in Pohang area(KME112, KME113, and KME114), statistical forecasting models of the daily maximum and mean values of PM10 have been developed. Since the distributions of the daily maximum and mean PM10 values are skewed, which are similar to the Weibull distribution, these values were log-transformed to increase prediction accuracy by approximating the normal distribution. Three statistical forecasting models, which are regression, neural networks(NN) and support vector regression(SVR), were built using the log-transformed response variables, i.e., log(max(PM10)) or log(mean (PM10)). Also, the forecasting models were validated by the measure of RMSE, CORR, and IOA for the model comparison and accuracy. The improvement rate of IOA before and after the log-transformation in the daily maximum PM10 prediction was 12.7% for the regression and 22.5% for NN. In particular, 42.7% was improved for SVR method. In the case of the daily mean PM10 prediction, IOA value was improved by 5.1% for regression, 6.5% for NN, and 6.3% for SVR method. As a conclusion, SVR method was found to be performed better than the other methods in the point of the model accuracy and fitness views.