• Journal of the Korea Institute of Military Science and Technology
• /
• v.13 no.4
• /
• pp.683-688
• /
• 2010

The ability to navigate autonomously in off-road terrain is the most critical technology needed for Unmanned Ground Vehicles(UGV). In this paper, we present a method for vision-based terrain cover classification using DCT features. To classify the terrain, we acquire image from a CCD sensor, then the image is divided into fixed size of blocks. And each block transformed into DCT image then extracts features which reflect frequency band characteristics. Neural network classifier is used to classify the features. The proposed method is validated and verified through many experiments and we compare it with wavelet feature based method. The results show that the proposed method is more efficiently classify the terrain-cover than wavelet feature based one.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.45 no.6
• /
• pp.55-59
• /
• 2008

For effective mobility control of UGV(unmanned ground vehicle), the terrain cover classification is an important component as well as terrain geometry recognition and obstacle detection. The vision based terrain cover classification algorithm consists of pre-processing, feature extraction, classification and post-processing. In this paper, we present a method to classify terrain covers based on the color and texture information. The color space conversion is performed for the pre-processing, the wavelet transform is applied for feature extraction, and the SVM(support vector machine) is applied for the classifier. Experimental results show that the proposed algorithm has a promising classification performance.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.4
• /
• pp.372-378
• /
• 2009

For effective mobility control of unmanned ground vehicles in outdoor off-road environments, terrain cover classification technology using passive sensors is vital. This paper presents a novel method far terrain classification based on color and texture information of off-road images. It uses a neural network classifier and wavelet features. We exploit the wavelet mean and energy features extracted from multi-channel wavelet transformed images and also utilize the terrain class spatial coordinates of images to include additional features. By comparing the classification performance according to applied features, the experimental results show that the proposed algorithm has a promising result and potential possibilities for autonomous navigation.

• Journal of Korea Multimedia Society
• /
• v.22 no.9
• /
• pp.1122-1131
• /
• 2019

This paper covers research on terrain construction for unmanned aerial vehicle simulators using spatial information that was distributed by public institutions. Aerial photography, DEM, vector maps and 3D model data were used in order to create a realistic terrain simulator. A data converting method was suggested while researching, so it was generated to automatically arrange and build city models (vWorld provided) and classification methods so that realistic images could be generated by 3D objects. For example: rivers, forests, roads, fields and so on, were arranged by aerial photographs, vector map (land cover map) and terrain construction based on the tile map used by DEM. In order to verify the terrain data of unmanned aircraft simulators produced by the proposed method, the location accuracy was verified by mounting onto Unreal Engine and checked location accuracy.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.853-854
• /
• 2008

The terrain perception technology using passive sensors plays a key role to enhance autonomous mobility for UGV. We present an effective method to classify terrain covers based on the color information. Considering a real-time implementation, neural network is applied for the terrain classifier and wavelet features extracted from the images are used. Test results show that the proposed algorithm has a promising classification performance.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.27 no.1
• /
• pp.115-127
• /
• 2024

This research aimed to construct models with various structures based on the Transformer module and to perform land cover classification, thereby examining the applicability of the Transformer module. For the classification of land cover, the Unet model, which has a CNN structure, was selected as the base model, and a total of four deep learning models were constructed by combining both the encoder and decoder parts with the Transformer module. During the training process of the deep learning models, the training was repeated 10 times under the same conditions to evaluate the generalization performance. The evaluation of the classification accuracy of the deep learning models showed that the Model D, which utilized the Transformer module in both the encoder and decoder structures, achieved the highest overall accuracy with an average of approximately 89.4% and a Kappa coefficient average of about 73.2%. In terms of training time, models based on CNN were the most efficient. however, the use of Transformer-based models resulted in an average improvement of 0.5% in classification accuracy based on the Kappa coefficient. It is considered necessary to refine the model by considering various variables such as adjusting hyperparameters and image patch sizes during the integration process with CNN models. A common issue identified in all models during the land cover classification process was the difficulty in detecting small-scale objects. To improve this misclassification phenomenon, it is deemed necessary to explore the use of high-resolution input data and integrate multidimensional data that includes terrain and texture information.

• Journal of Korean Society for Geospatial Information Science
• /
• v.18 no.2
• /
• pp.29-33
• /
• 2010

The purpose of LiDAR ground classification is to archive both goals which are acquiring confident ground points with high precision and describing ground shape in detail. In spite of many studies about developing optimized algorithms to kick out this, it is very difficult to classify ground points and describing ground shape by airborne LiDAR data. Especially it is more difficult in a dense forested area like Korea. Principle misclassification was mainly caused by complex forest canopy hierarchy in Korea and relatively coarse LiDAR points density for ground classification. Unfortunately, a lot of LiDAR surveying performed in summer in South Korea. And by that reason, schematic LiDAR points distribution is very different from those of Europe. So, this study propose enhanced ground classification method considering Korean land cover characteristics. Firstly, this study designate highly confident candidated LiDAR points as a first ground points which is acquired by using big roller classification algorithm. Secondly, this study applied weighted gradient kernel(WGK) algorithm to find and include highly expected ground points from the remained candidate points. This study methods is very useful for reconstruct deformed terrain due to misclassification results by detecting and include important terrain model key points for describing ground shape at site. Especially in the case of deformed bank side of river area, this study showed highly enhanced classification and reconstruction results by using WGK algorithm.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2006.04a
• /
• pp.439-444
• /
• 2006

Korea Peninsula consists of approximately 70% of mountainous terrain of total area, in addition, annual average rainfall is plentiful, especially during rainy season of summer, and it is often accompanied with typhoon and heavy rain, which results in frequent landslides. Since there are limitations with existing methods to analyze extensive disasters, it is necessary to develop new remote sensing technology using an artificial satellite to study on landslides closely. This paper is written in order to establish the database with map information on various landslides using GIS, furthermore, to analyze precariousness of the areas, which are susceptible to landslide, and risks of potential areas in consideration of heavy rain, based on land-cover classification derived from images from satellite.

• Journal of the Korean Geographical Society
• /
• v.38 no.3
• /
• pp.375-388
• /
• 2003

This study mainly focuses on evaluating how the triangular training operator could improve classification accuracy in SAR(Synthetic Aperture Radar) and VIR FCC(Visible Infra-red, False Colour Composite). The techniques for the determination of the most informative SAR/VIR combinations in the triangular space diagram, as developed tv the author of the paper, are given and the results obtained are presented. The SAR alone, VIR alone and SAR/VIR FCC classification showed trends for gradual improvement of accuracy. Accuracy distribution pattern for individual classes could be explained closely related to SAR/VIR signature components in the process of the triangular synergistic training. Due to contribution of SAR signature in training samples, it was possible to isolate major terrain features such as cloud cover area and roughness target with acceptable spatial precision. It is anticipated that this research output could be used as a valuable reference for distribution trends of classification accuracy obtained by triangular channel space based training in synergistic application.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.13 no.1
• /
• pp.43-66
• /
• 1985

Korea needs proper and rational public policies on conservation and use of forest land and other natural resources because of the accelerating expansion of national land developments in recent years. Unfortunately, there is no systematic planning system to support the needs. Generally, forest land use planning needs suitability analysis based on efficient land classification system. The goal of this study was to classify a forest land using multivariate satistical methods. A case study was carried out in winter of 1983 on a mountainous area higher than 100m above sea level located at Mt. Kwanak in Anyang -city, Kyung-gi-do (province). The study area was 19.80 km$^2$wide and was divided into 1, 383 Operational Taxonomic Units (OTU's) by a 120m$\times$120m grid. Fourteen descriptors were identified and quantified for each OTU from existing national land data : elevation, slope, aspect, terrain form, geologic material, surface soil permeability, topsoil type, depth of the solum, soil acidity, forest cover type, stand size class, stand age class, stand density class, and simple forest soil capability class. For this study, a FORTRAN IV program was written for input and output map data, and the computer statistics packages, SPSS and BMD, were used to perform the multivariate statistical analysis. Fourteen variables were analyzed to investigate the characteristics of their fire quench distribution and to estimate the correlation coefficients among them. Principal component analysis was executed to find the dimensions of forest land characteristics, and factor scores were used for proper samples of OTU throughout the study area. In order to develop the classes of forest land classification based on 102 surrogates, cluster and discriminant analyses of principal descriptor variable matrix were undertaken. Results obtained through a series of multivariate statistical analyses were as follows ; 1) Principal component analysis was proved to be a useful tool for data selection and identification of principal descriptor variables which represented the characteristics of forest land and facilitated the selection of samples.