• Tunnel and Underground Space
• /
• v.29 no.4
• /
• pp.230-242
• /
• 2019

While various automatic rock fracture survey methods have been researched, the evaluation of the accuracy of these methods raises issues due to the absence of a metric which fully expresses the similarity between automatic and manual fracture maps. Therefore, this paper proposes a geometry similarity metric which is especially designed to determine the overall similarity of fracture maps and to evaluate the accuracy of rock fracture survey methods by a single number. The proposed metric, Scanline Intersection Similarity (SIS), is derived by conducting a large number of scanline surveys upon two fracture maps using Python code. By comparing the frequency of intersections over a large number of scanlines, SIS is able to express the overall similarity between two fracture maps. The proposed metric was compared with Intersection Over Union (IoU) which is a widely used evaluation metric in computer vision. Results showed that IoU is inappropriate for evaluating the geometry similarity of fracture maps because it is overly sensitive to minor geometry differences of thin elongated objects. The proposed metric, on the other hand, reflected macro-geometry differences rather than micro-geometry differences, showing good agreement with human perception. The metric was further applied to evaluate the accuracy of a deep learning-based automatic fracture surveying method which resulted as 0.674 (SIS). However, the proposed metric is currently limited to 2D fracture maps and requires comparison with rock joint parameters such as RQD.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 1999.05a
• /
• pp.50-54
• /
• 1999

This paper is proposed to image retrieval system using color and shape feature. Color feature used to four maximum value feature among the maximum value extracted from local color distribution histogram. The preprocessing of shape feature consist of edge extraction and weight central point extraction and angular sampling. The sum of distance from weight central point to contour and variation and max/min used to shape feature. The similarity is estimated compare feature of query image with the feature of images in database and the candidate of image is retrieved in order of similarity. We evaluate the effectiveness of shape feature and color feature in experiment used to two hundred of the closed image. The Recall and the Precision is each 0.72 and 0.53 in the result of average experiment. So the proposed method is presented useful method.

• Journal of Korean Society of Forest Science
• /
• v.103 no.1
• /
• pp.12-22
• /
• 2014

This study was carried out to classify forest communities and to aggregate forest cover types for the complex and diversified natural forest areas of Hwangaksan, Bakseoksan, Deogyusan, and Jirisan in southern region of Baekdudaegan Mountains. The vegetation data were collected by point-centered quarter sampling method. Eight hundred fifty one sample points were subjected to cluster analysis to classify 18 forest communities, which were aggregated into 7 representative forest cover types on the basis of community similarity from composition of canopy species. They were mixed mesophytic forest cover type, the others deciduous forest cover type, Quercus variabilis-Quercus serrata cover type, Quercus mongolica cover type, Pinus densiflora cover type, Carpinus laxiflora cover type, and Abies koreana cover type. The Quercus mongolica cover type was most widely distributed in the study areas, and this cover type tended to occur in the place of higher altitude as latitude was getting lower. Mixed mesophytic forest and the others deciduous forest cover type were commonly distributed in the areas of valley, on the other hand, Quercus mongolica cover type and Pinus densiflora cover type tended to be distributed in the areas of ridge.

• Korean Journal of Environment and Ecology
• /
• v.29 no.5
• /
• pp.729-735
• /
• 2015

This study was carried out to analyze the community structure and distribution of macrobenthos in Hakampo sandy shore. As for the environmental factors of sampling stations, the average grain size of the sediments from the sampling stations was $2.37{\phi}$ with the range of $1.90{\phi}{\sim}2.52{\phi}$, which represents the typical medium sand sedimentary facies of sandy beach. The average sediment organic content was 1.15% with the range of 0.90%~1.46%. Most stations had low sediment organic content which is typical of sandy beach. The number of species were total 59 from 6 taxa and the individuals was $668inds./m^2$ in the 10 stations of the study area. Based on the Bray-Curtis similarity analysis, the 10 stations of the study area were broadly divided into 3 groups. The highest part of intertidal zone was St. 2~St. 4, the middle part of intertidal zone was St. 5~St. 8, and the lower part of intertidal zone was St. 9~St. 10. A spatial community structure was observed which is divided into the highest, middle, and lower parts.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.26 no.2
• /
• pp.172-178
• /
• 2016

The location template matching (LTM) method is a technique of identifying an impact location on a structure, and is often applied to structural health monitoring and large scale human-computer interface (HCI) systems. The LTM method utilizes a certain measure of similarity between two time signals. The correlation coefficient is most widely used for this purpose, and the group delay based method is recently proposed to improve the accuracy of finding the best matching pair of signals. In practice, one of key essential consideration for implementing the LTM method is to guarantee that a sufficient number of vibration modes must be contained in the measured signal, and yet the lower sampling rate is needed for a real-time implementation. In this paper, the properties of correlation coefficient and group delay with respect to the number of vibration modes are investigated. A few important results are obtained through extensive computer simulations and experiments. If the number of vibration modes contained in the measured signal is more than four it is sufficient for the correlation based LTM method, while the group delay based LTM method requires smaller number of vibration modes.

• Journal of Korean Society for Atmospheric Environment
• /
• v.21 no.E3
• /
• pp.107-113
• /
• 2005

Using a sampling device of our own making, airborne pollen has been monitored in Kyoto, Japan from the middle of February to the end of May 2004. From the morphological analysis of pollen grains by Scanning Electron Microscope (SEM), it was possible to identify some pollen types like Cryptomeria, Pine, Alder, Cyclobalanopsis, Chamaecyparis, and Equisetum. Daily average airborne pollen counts show strong variations from the day to day which makes the appropriate daily forecasts that could be of practical use for patients difficult. Diurnal variation of airborne pollen grains at our local sampling site is very irregular and shows no similarity between pollen types. The highest concentrations of Cryptomeria and Alder pollens in the south -west wind directions might be attributed to the airborne pollen transport, while the increase in Pine pollen grain in the southern wind direction was probably due to the local spread. Prevailing wind direction (SW) during the pollinating periods of Cryptomeria and Alder pollens could suggest a long-distance transport from a distant mountain.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.7
• /
• pp.83-90
• /
• 2008

EThe digitalized image information has various resolution, and it has been developed for several technologies to resize image depending on user's requests and applications. Recently the algorithm using edge information for good image/video quality on up-sampling was introduced, and the pre-processing procedure is required for edge extraction. Than, the predicted direction in intra prediction used in H.264/AVC has the similarity up to 80% for the edge information, so I propose the image up-sampling method using edge information. In proposed method, the image quality is similar to result of adapting n'th kernel. and the method reduces the number of calculation by about 50%.

• Journal of Fisheries and Marine Sciences Education
• /
• v.29 no.2
• /
• pp.526-536
• /
• 2017

This study intended to understand similarity of image between games and position of image attribute perceived by consumers according to the games of marine sports using MDS (Multidimensional Scaling). Through the foregoing, this study aims to provide preliminary data for establishing marketing strategies for games of marine sports by accurately understanding images of marines sports perceived by consumers. For survey targets, this study selected students of K University located in Gyeonggi-do as a population, and extracted samples using convenience sampling out of non-probability sampling methods targeting 200 students who showed intention of participation in this study, and total 188 sheets of questionnaire were used as final data excepting 12 sheets that are filled up unfaithfully or considered unreliable. For data processing, this study conducted Frequency Analysis, Descriptive Statistical Analysis, Reliability Analysis, MDS (Multidimensional Scaling) and Multiple Regression Analysis, and study results show that water ski and wakeboard(.602) have the most similar image attribute, indicating that image attributes of scuba diving and water ski(2.031) are positioned farthest with each other. As for image attribute, image attribute of water ski has appearance and progressiveness, windsurfing has an image attribute of positive, scuba diving has an image attribute of negation, and marin rafting has an image attribute of friendliness.

• The Journal of Asian Finance, Economics and Business
• /
• v.8 no.3
• /
• pp.1385-1393
• /
• 2021

This study examines the impact of independent behavior, goal achievement, and curiosity on employee performance and examines the impact of independent behavior and goal achievement on employee performance through curiosity. This research is conducted on a survey using explanatory research. Data collection is carried out using a questionnaire as a research instrument. The questionnaire was conducted on employees of finance companies in Yogyakarta, Indonesia. The selection of companies used as research is based on the similarity of the products being marketed. In Yogyakarta, there are 54 branch offices of finance companies that carry out business activities. However, of the 54 companies, ten have the same products being marketed, namely mortgage loans, investment loans, and multipurpose loans. In contrast, other finance companies focus more on motor vehicle loans—determining the number of samples obtained by multiplying by six times the number of indicators used. The sampling technique is done by purposive sampling with specific respondent criteria. In this study, analysis techniques, using Structural Equation Modeling (SEM) with AMOS 22. The results showed that the increase in employee performance was influenced by curiosity regarding independent employee behavior. This study provides recommendations human resource management practices that have an impact on organizational performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.11
• /
• pp.4184-4202
• /
• 2021

Text steganography combined with natural language generation has become increasingly popular. The existing methods usually embed secret information in the generated word by controlling the sampling in the process of text generation. A candidate pool will be constructed by greedy strategy, and only the words with high probability will be encoded, which damages the statistical law of the texts and seriously affects the security of steganography. In order to reduce the influence of the candidate pool on the statistical imperceptibility of steganography, we propose a steganography method based on a new sampling strategy. Instead of just consisting of words with high probability, we select words with relatively small difference from the actual sample of the language model to build a candidate pool, thus keeping consistency with the probability distribution of the language model. What's more, we encode the candidate words according to their probability similarity with the target word, which can further maintain the probability distribution. Experimental results show that the proposed method can outperform the state-of-the-art steganographic methods in terms of security performance.