• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.14 no.4
• /
• pp.313-321
• /
• 2014

In the current era of data-intensive services, the handling of big data is a crucial issue that affects almost every discipline and industry. In this study, we propose a classification method for large volumes of numeric data, which is implemented in a distributed programming framework, i.e., MapReduce. The proposed method partitions the data space into a grid structure and it then models the probability distributions of classes for grid cells by collecting sufficient statistics using distributed MapReduce tasks. The class labeling of new data is achieved by k-nearest neighbor classification based on Bayesian inference.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.1
• /
• pp.364-380
• /
• 2016

The most popular approach in object classification is based on the bag of visual-words model, which has several fundamental problems that restricting the performance of this method, such as low time efficiency, the synonym and polysemy of visual words, and the lack of spatial information between visual words. In view of this, an object classification based on weakly supervised E2LSH and saliency map weighting is proposed. Firstly, E2LSH (Exact Euclidean Locality Sensitive Hashing) is employed to generate a group of weakly randomized visual dictionary by clustering SIFT features of the training dataset, and the selecting process of hash functions is effectively supervised inspired by the random forest ideas to reduce the randomcity of E2LSH. Secondly, graph-based visual saliency (GBVS) algorithm is applied to detect the saliency map of different images and weight the visual words according to the saliency prior. Finally, saliency map weighted visual language model is carried out to accomplish object classification. Experimental results datasets of Pascal 2007 and Caltech-256 indicate that the distinguishability of objects is effectively improved and our method is superior to the state-of-the-art object classification methods.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.763-765
• /
• 2003

We made a hazard map using GIS and remote sensing for he greatest inundation damage that happened for the 20th century. We calculated the land cover classification using Landsat from 1983 to 2000. We calculated it from a damage report and an aerial photo for a flood. We considered relation of both land cover classification and the damage. We expected the inundation damage in the future and made a hazard map.

• Korean Journal of Remote Sensing
• /
• v.32 no.2
• /
• pp.105-118
• /
• 2016

The purpose of this study is to improve the classification system of sub-divided land cover map among the land cover maps provided by the Ministry of Environment. To accomplish the purpose, first, the overseas country land cover map classification items were examined in priority. Second, the area ratio of each item established by applying the previous sub-divided classification system was analyzed. Third, the survey on the improvement of classification system targeting the users (experts and general public) who actually used the sub-divided land cover map was carried out. Fourth, a new classification system which improved the previous system by reclassifying 41 classification items into 33 items was finally established. Fifth, the established land cover classification items were applied on study area, and the land cover classification result according to the improvement method was compared with the previous classification system. Ilsan area in Goyang city where there are diverse geographic features with various land surface characteristics such as the urbanization area and agricultural land were distributed evenly were selected as the study area. The basic images used in this study were 0.25 m aerial ortho-photographs captured by the National Geographic Information Institute (NGII), and digital topographic map, detailed stock map plan, land registration map and administrative area map were used as the relevant reference data. As a result of applying the improved classification system into the study area, the area of culture-sports, leisure facilities was $1.84km^2$ which was approximately more than twice larger in comparison to the previous classification system. Other areas such as transportation and communication system and educational administration facilities were not classified. The result of this study has meaningful significance that it reflects the efficiency for the establishment and renewal of sub-divided land cover map in the future and actual users' needs.

• The Korean Journal of Ecology
• /
• v.19 no.5
• /
• pp.465-476
• /
• 1996

There are several environmental variables that may be influential to the spatial distribution of forest vegetation. To create a map of forest vegetation zone over southern part of Korean Peninsula, digital map layers were produced for each of environmental variables that include topography, geographic locations, and climate. In addition, an extensive set of field survey data was collected at relatively undisturbed forests and they were introduced into the GIS database with exact coordinates of survey sites. Preliminary statistical analysis on the survey data showed that the environmental variables were significantly different among the previously defined five forest vegetation zones. Classification of the six layers of digital map representing environmental variables was carried out by a supervised classifier using the training statistics from field survey data and by a clustering algorithm. Although the maps from two classifiers were somewhat different due to the classification procedure applied, they showed overall patterns of vertical and horizontal distribution of forest zones. considering the spatial contents of many ecological studies, GIS can be used as an important tool to manage and analyze spatial data. This study discusses more about the generation of digital map and the analysis procedure rather than the outcome map of forest vegetation zone.

• IEIE Transactions on Smart Processing and Computing
• /
• v.5 no.1
• /
• pp.43-48
• /
• 2016

We propose plain fingerprint classification based on a core stochastic algorithm that effectively uses a core stochastic model, acquiring more fingerprint minutiae and direction, in order to increase matching performance. The proposed core stochastic algorithm uses core presence/absence and contains a ridge direction and distribution map. Simulations show that the fingerprint classification accuracy is improved by more than 14%, on average, compared to other algorithms.

• Aerospace Engineering and Technology
• /
• v.6 no.1
• /
• pp.213-221
• /
• 2007

This paper showed how to analysis the object-based classification for wildfire fuel type map using Hyperion hyperspectral remote sensing data acquired in April, 2002 and compared the results of the object-based classification with the results of the pixel-based classification. Our methodological approach for wildfire fuel type map firstly processed correcting abnormal pixels and atypical bands and also calibrating atmospheric noise for enhanced image quality. Fuel type map is characterized by the results of the spectral mixture analysis(SMA). Object-based approach was based on segment-based endmember selection, while pixel-based method used standard SMA. To validate and compare, we used true-color high resolution orthoimagery.

• Spatial Information Research
• /
• v.11 no.2
• /
• pp.101-117
• /
• 2003

High resolution satellite image analysis has been recognized as an effective technique for monitoring local land-cover and atmospheric changes. In this study, a new high resolution map for land-cover was generated using both high-resolution IKONOS image and conventional land-use mapping. Fuzzy classification method was applied to classify land-cover, with minimum operator used as a tool for joint membership functions. In separateness analysis, the values were not great for all bands due to discrepancies in spectral reflectance by seasonal variation. The land-cover map generated in this study revealed that conifer forests and farm land in the ground and tidal flat and beach in the ocean were highly changeable. The kappa coefficient was 0.94% and the overall accuracy of classification was 95.0%, thus suggesting a overall high classification accuracy. Accuracy of classification in each class was generally over 90%, whereas low classification accuracy was obtained for classes of mixed forest, river and reservoir. This may be a result of the changes in classification, e.g. reclassification of paddy field as water area after water storage or mixed use of several classification class due to similar spectral patterns. Seasonal factors should be considered to achieve higher accuracy in classification class. In conclusion, firstly, IKONOS image are used to generated a new improved high resolution land-cover map. Secondly, IKONOS image could serve as useful complementary data for decision making when combined with GIS spatial data to produce land-use map.

• The Journal of the Acoustical Society of Korea
• /
• v.16 no.2
• /
• pp.58-65
• /
• 1997

A new automatic classification method with high and stable accuracy for time-series image data is presented in this paper. This method is based on prior condition that a classified map of the target area already exists, or at least one of the time-series image data had been classified. The classified map is used as a reference map to specify training areas of classification categories. The new automatic classification method consists of five steps, i.e., extraction of training data using reference map, detection of changed pixels based upon the homogeneity of training data, clustering of changed pixels, reconstruction of training data, and classification as like maximum likelihood classifier. In order to evaluate the performance of this method qualitatively, four time-series Landsat TM image data were classified by using this method and a conventional method which needs a skilled operator. As a results, we could get classified maps with high reliability and fast throughput, without a skilled operator.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.11 no.2
• /
• pp.1-9
• /
• 2008

The objective of the study is 10 know the relation of landslide occurrence with using TPI (Topographic Position Index) in the Pyungchang County. Total 659 landslide scars were detected from aerial photographs. To analyze TPI, 100m SN (Small-Neighborhood) TPI map, 500m LN (Large-Neighborhood) TPI map, and slope map were generated from the DEM (Digital Elevation Model) data which are made from 1 : 5,000 digital topographic map. 10 classes clustered by regular condition after overlapping each TPI maps and slope map. Through this process, we could make landform classification map. Because it is only to classify landform, 7 classes were finally regrouped by the slope angle information of landslide occurrence detected from aerial photography analysis. The accuracy of reclassified map is about 46%.