• The KIPS Transactions:PartB
• /
• v.12B no.4 s.100
• /
• pp.429-436
• /
• 2005

A simple stereo matching algorithm using population-based incremental learning(PBIL) is proposed in this paper to decrease the general problem of genetic algorithms, such as memory consumption and inefficiency of search. PBIL is a variation of genetic algorithms using stochastic search and competitive teaming based on a probability vector. The structure of PBIL is simpler than that of other genetic algorithm families, such as serial and parallel ones, due to the use of a probability vector. The PBIL strategy is simplified and adapted for stereo matching circumstances. Thus, gene pool, chromosome crossover, and gene mutation we removed, while the evolution rule, that fitter chromosomes should have higher survival probabilities, is preserved. As a result, memory space is decreased, matching rules are simplified and computation cost is reduced. In addition, a scheme controlling the distance of neighbors for disparity smoothness is inserted to obtain a wide-area consistency of disparities, like a result of coarse-to-fine matchers. Because of this scheme, the proposed algorithm can produce a stable disparity map with a small fixed-size window. Finally, an alterative version of the proposed algorithm without using probability vector is also presented for simpler set-ups.

• Journal of Korean Society of Rural Planning
• /
• v.16 no.2
• /
• pp.47-55
• /
• 2010

Land cover and land use change data are important in many studies including climate change and hydrological studies. Although the various theories and models have been developed, it is difficult to identify the driving factors of the land use change because land use change is related to policy options and natural and socio-economic conditions. This study is to attempt to simulate the land cover change using the CLUE model based on a statistical analysis of land-use change. CLUE model has dynamic modeling tools from the competition among land use change in between driving force and land use, so that this model depends on statistical relations between land use change and driving factors. In this study, Yongin, Icheon and Anseong were selected for the study areas, and binary logistic regression and factor analysis were performed verifying with ROC curve. Land cover probability map was also prepared to compare with the land cover data and higher probability areas are well matched with the present land cover demonstrating CLUE model applicability.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.13 no.1
• /
• pp.74-88
• /
• 2010

In this study, the points of LiDAR were modified in order to generate various DEM resolutions by applying LiDAR data in Ulsan. Since the LiDAR data have points with 1m intervals, the number of points for each resolution was modified to the size of 1, 5, 10, 30, 50, 100m by uniformly eliminating the points. A runoff analysis was performed on Taehwa river and its tributary, Dongcheon, with 200 year rainfall exceedance probability. 2-dimensional inundation analysis was performed based on the density of LiDAR data using FLUMEN, which was used to establish domestic flood risk map. Once DEM data obtained from LiDAR survey are used, it is expected that the study results can be used as data in determining optimal grid spacing, which is economical, effective and accurate in establishing flood defence plans including the creation of flood risk map.

• The Korean Journal of Ecology
• /
• v.24 no.6
• /
• pp.385-394
• /
• 2001

An attempt to clarify the landscape structure of urban areas was carried out in the greenbelt around Seoul, Korea's metropolis. By means of aerial photographs and a field survey, a vegetation map including land-use pattern was made. Landscape structure was described by analyzing this vegetation map and the results of phytosociological survey. Landscape element types identified were (1) secondary forest, (2) plantation, (3) cultivated field, (4) urbanized area, (5) graveyard, and (6) bare rock. Vegetation units, resulting from the phytosociological analysis, included Quercus mongolica, Q. variabilis, Q. acutissima, Pinus densiflora, Q. aliena, and Alnus japonica communities. Plantations were composed of Robinia pseudoacacia, Populus tomentiglandulosa, P. rigida, Larix leptolepis, P. koraiensis, and Castanea crenata stands. Patches near to human settlements in the lower zones of the mountains were fragmented and small but they became larger towards the higher mountain zones. On the other hand, the number of patches was fewer and their size was larger in Mt. Cheonggye more distant from the principal residential area, larger in size, and higher in elevation compared with the other 2 mountains, Mt. Daemo and Mt. Acha. Floristic composition of Mongolian oak(Q. mongolica) stand distributing in the upper part of each mountain, in which artificial interference is rare, showed a difference among those study areas different in parent rock and disturbance regime. But that of black locust(R. pseudoacacia) stand located in lowland of mountainous area, in which artificial interference is frequent was similar to each other. As the results of analyses on the frequency distribution of diameter classes of major species, dominant landscape elements, Mongolian oak forest showed different responses depending on artificial interference as continuous maintenance and retrogressive succession in the sites far from and near to the residential areas, respectively. On the other hand, black locust stands showed a probability to be restore to the native oak forest through progressive succession.

• Journal of Korea Multimedia Society
• /
• v.14 no.2
• /
• pp.288-294
• /
• 2011

In this paper we define the probability models for determining the disparity map given stereo images and derive the methods for solving the problem, which is proven to be equivalent to an energy-based stereo matching. Under the assumptions the difference between the pixel on the left image and the corresponding pixel on the right image and the difference between the disparities of the neighboring pixels are exponentially distributed, a recursive approach for estimating the MRF regularizing parameter is proposed. The proposed method alternates between estimating the parameters with the intermediate disparity map and estimating the disparity map with the estimated parameters, after computing it with random initial parameters. Our algorithm is applied to the stereo matching algorithms based on the dynamic programming and belief propagation to verify its operation and measure its performance.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.1
• /
• pp.112-119
• /
• 2010

This paper defines the probability models for determining the disparity map given stereo images and derives the methods for solving the problem, which is proven to be equivalent to an energy-based stereo matching. Under the assumptions the difference between the pixel on the left image and the corresponding pixel on the right image and the difference between the disparities of the neighboring pixels are exponentially distributed, a recursive approach for estimating the MRF regularizing parameter is proposed. Usually energy-based stereo matching methods are so sensitive to the parameter that it should be carefully determined. The proposed method alternates between estimating the parameter with the intermediate disparity map and estimating the disparity map with the estimated parameter, after computing it with random initial parameter. It is shown that the parameter estimated by the proposed method converges to the optimum and its performance can be improved significantly by adjusting the parameter and modifying the energy term.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.2
• /
• pp.397-404
• /
• 2018

As the usage of internet is increasing, a lot of user behavior are written in a log file and the researches and industries using the log files are getting activated recently. This paper uses the Hadoop based on open source distributed computing platform and proposes a customized tourism system by analyzing user behaviors in the log files. The proposed system uses Google Analytics to get user's log files from the website that users visit, and stores search terms extracted by MapReduce to HDFS. Also it gathers features about the sight-seeing places or cities which travelers want to tour from travel guide websites by Octopus application. It suggests the customized cities by matching the search terms and city features. NBP(next bit permutation) algorithm to rearrange the search terms and city features is used to increase the probability of matching. Some customized cities are suggested by analyzing log files for 39 users to show the performance of the proposed system.

• Journal of Korean Society for Geospatial Information Science
• /
• v.19 no.3
• /
• pp.33-40
• /
• 2011

This paper presents the prediction and evaluation of landslide using LRA(logistic regression analysis) and ANN (Artificial Neural Network) methods. In order to assess the landslide, we selected Sarabong, Byeoldobong area and Mt. Song-ak in Jeju Island. Five factors which affect the landslide were selected as: slope angle, elevation, porosity, dry density, permeability. So as to predict and evaluate the landslide, firstly the weight value of each factor was analyzed by LRA(logistic regression analysis) and ANN(Artificial Neural Network) methods. Then we got two prediction maps using AcrView software through GIS(Geographic Information System) method. The comparative analysis reveals that the slope angle and porosity play important roles in landslide. Prediction map generated by LRA method is more accurate than ANN method in Jeju. From the prediction map, we found that the most dangerous area is distributed around the road and path.

• Journal of The Geomorphological Association of Korea
• /
• v.27 no.3
• /
• pp.87-103
• /
• 2020

The purpose of this study is to evaluate the risk of cropland and man-made infrastructures in a landslide-prone area using a GIS-based method. To achieve this goal, a landslide inventory map was prepared based on aerial photograph analysis as well as field observations. A total of 550 landslides have been counted in the entire study area. For model analysis and validation, extracted landslides were randomly selected and divided into two groups. The landslide causative factors such as slope, aspect, curvature, topographic wetness index, elevation, forest type, forest crown density, geology, land-use, soil drainage, and soil texture were used in the analysis. Moreover, to identify the correlation between landslides and causative factors, pixels were divided into several classes and frequency ratio was also extracted. A landslide susceptibility map was constructed using a bayesian predictive model (BPM) based on the entire events. In the cross validation process, the landslide susceptibility map as well as observation data were plotted with a receiver operating characteristic (ROC) curve then the area under the curve (AUC) was calculated and tried to extract a success rate curve. The results showed that, the BPM produced 85.8% accuracy. We believed that the model was acceptable for the landslide susceptibility analysis of the study area. In addition, for risk assessment, monetary value (local) and vulnerability scale were added for each social thematic data layers, which were then converted into US dollar considering landslide occurrence time. Moreover, the total number of the study area pixels and predictive landslide affected pixels were considered for making a probability table. Matching with the affected number, 5,000 landslide pixels were assumed to run for final calculation. Based on the result, cropland showed the estimated total risk as US $ 35.4 million and man-made infrastructure risk amounted to US $ 39.3 million.

• Journal of Practical Engineering Education
• /
• v.15 no.2
• /
• pp.509-517
• /
• 2023

The reactive web service provides a related word recommendation system using the TextRank algorithm and a word-based idea generation service selected by the user. In the related word recommendation system, the method of weighting each word using the TextRank algorithm and the probability output method using SoftMax are discussed. The idea generation service discusses the idea generation method and the artificial intelligence reinforce-learning method using mini-GPT. The reactive web discusses the linkage process between React, Spring Boot, and Flask, and describes the overall operation method. When the user enters the desired topic, it provides the associated word. The user constructs a mind map by selecting a related word or adding a desired word. When a user selects a word to combine from a constructed mind-map, it provides newly generated ideas and related patents. This web service can share generated ideas with other users, and improves artificial intelligence by receiving user feedback as a horoscope.