• Korean Journal of Remote Sensing
• /
• v.36 no.6_2
• /
• pp.1579-1590
• /
• 2020

In this paper, spatially-adaptive denormalization (SPADE) based U-Net is proposed to detect changes by using high-resolution satellite images. The proposed network is to preserve spatial information using SPADE. Change detection methods using high-resolution satellite images can be used to resolve various urban problems such as city planning and forecasting. For using pixel-based change detection, which is a conventional method such as Iteratively Reweighted-Multivariate Alteration Detection (IR-MAD), unchanged areas will be detected as changing areas because changes in pixels are sensitive to the state of the environment such as seasonal changes between images. Therefore, in this paper, to precisely detect the changes of the objects that consist of the city in time-series satellite images, the semantic spatial objects that consist of the city are defined, extracted through deep learning based image segmentation, and then analyzed the changes between areas to carry out change detection. The semantic objects for analyzing changes were defined as six classes: building, road, farmland, vinyl house, forest area, and waterside area. Each network model learned with KOMPSAT-3A satellite images performs a change detection for the time-series KOMPSAT-3 satellite images. For objective assessments for change detection, we use F1-score, kappa. We found that the proposed method gives a better performance compared to U-Net and UNet++ by achieving an average F1-score of 0.77, kappa of 77.29.

• Korean Journal of Agricultural Science
• /
• v.38 no.1
• /
• pp.109-113
• /
• 2011

The purpose of this study was to examine the potentiality of urban forest roads as an environment for enhancing physical fitness. Six male university students participated in the study as subjects. The subjects walked on an urban forest road for 30 minutes. As a control experiment, they also walked on a national park trail for 30 minutes. Subjects' heart rates were monitored during the walks to calculate the ratio of the average time their heart rates were within the target range (from 60% to 80% of the maximal heart rate) for Enhancing Physical Fitness. After the walks, images of the spaces were analyzed using the semantic differential (SD) method. During the walk on the urban forest road, subjects' heart rates were within the target range 63.3% of the time, and lower than the target range 36.7% of the time. During the control experiment on the national park trail, subjects' heart rates were within the target range only 23.3% of the time, and higher than the target range 76.7% of the time. From the spatial perception evaluation using the SD method, subjects' comfortable and natural feelings when they were on the national park trail were significantly greater than when they were on the urban forest trail, but there were no differences in terms of other SD descriptors, such as friendliness and likeability. The results of our study indicate that the urban forest road provides a good environment for walking to enhance physical fitness. Although not as close to nature as national park trails, urban forest roads offer similar natural environments and have a high potentiality for serving as leisure spaces for urban residents who seek physical activities.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.9
• /
• pp.53-62
• /
• 2013

This paper proposes a method of automatically predicting the vanishing point for the purpose of detecting the road region from natural images. The proposed method stably detects the vanishing point in the road environment by analyzing the dominant orientation of the image and predicting the vanishing point to be at the position where the feature components of the image are concentrated. For this purpose, in the first stage, the image is partitioned into sub-blocks, an edge sample is selected randomly from within the sub-block, and RANSAC is applied for line fitting in order to analyze the dominant orientation of each sub-block. Once the dominant orientation has been detected for all blocks, we proceed to the second stage and randomly select line samples and apply RANSAC to perform the fitting of the intersection point, then measure the cost of the intersection model arising from each line and we predict the vanishing point to be located at the average point, based on the intersection point model with the highest cost. Lastly, quantitative and qualitative analyses are performed to verify the performance in various situations and prove the efficiency of the proposed algorithm for detecting the vanishing point.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.13 no.6
• /
• pp.1-9
• /
• 2014

Recently, a study is conducting to image collection and auto detection of attribute information using mobile mapping system. The road sign attribute information detection is difficult because of various size and placement, interference of other facilities like trees. In this study, a text detection method that does not rely on a Korean character template is required to successfully detect the target text when a variety of differently sized texts are present near the target texts. To overcome this, the method of incremental right-to-left blob projection (IRBP) was suggested as a solution; the potential and improvement of the method was also assessed. To assess the performance improvement of the IRBP that was developed, the IRBP method was compared to the existing method that uses Korean templates through the 60 videos of street signs that were used. It was verified that text detection can be improved with the IRBP method.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.17 no.1
• /
• pp.129-136
• /
• 2018

In this paper, we propose a new real-time lane detection method that is efficient for road environment. Existing methods have a problem of low reliability under environmental changes. In order to overcome this problem, we emphasize the lane candidate area by using gray level division. And Extracts a straight line component near the lane by using the Hough transform, and generates an ROI for each straight line based on the extracted coordinates. And integrates the generated ROI images. Then, the lane is determined by dividing the object using the dual queue in the ROI image. The proposed method is able to detect lanes even in the environmental change unlike the conventional method. And It is possible to obtain an advantage that the area corresponding to the background such as sky, mountain, etc. is efficiently removed and high reliability is obtained.

• Journal of Korean Society for Geospatial Information Science
• /
• v.13 no.4 s.34
• /
• pp.3-9
• /
• 2005

The extraction of traffic information based on image processing is under broad research recently because the method based on image processing takes less cost and effort than the traditional method based on physical equipment. The main purpose of the algorithm based on image processing is to extract vehicles from an image correctly. Before the extraction, the algorithm needs the pre-processing such as background subtraction and binary image thresholding. During the pre-processing much noise is brought about because roadside tree and passengers in the sidewalk as well as vehicles are extracted as traffic flow. The noise undermines the overall accuracy of the algorithm. In this research, most of the noise could be removed by extracting the exact road area which does not include sidewalk or roadside tree. To extract the exact road area, traffic lanes in the image were used. Algorithm speed also increased. In addition, with the ratio between the sequential images, the problem caused by vehicles' shadow was minimized.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.12 no.5
• /
• pp.162-170
• /
• 2004

In these days, a vehicle simulator is developed by using a VR(Virtual Reality) system. A VR system must provide a vehicle simulator with a natural interaction, a sufficient immersion and realistic images. To achieve this, it is important to provide a fast and uniform rendering performance regardless of the complexity of virtual worlds or the level of simulation. In this paper, modeling methods which offer an improved rendering performance for complex VR applications as 3D road model have been implemented and verified. The key idea of the methods is to reduce a load of VR system by means of LOD(Level of Detail), alpha blending texture mapping, texture mip-mapping and bilboard. Hence, in 3D road model where a simulation is complex or a scene is very large, the methods can provide uniform and acceptable frame rates. The VR system which is constructed with the methods has been experimented under the various application environments. It is confirmed that the proposed methods are effective and adequate to the VR system which associates with a vehicle simulator.

• Journal of the Institute of Convergence Signal Processing
• /
• v.8 no.2
• /
• pp.98-105
• /
• 2007

In this paper, we propose the method and system which determines the condition for safe and unsafe lane changing. To determine the condition, first, the system sets up the Region of Interest(ROI) on the neighboring lane. Second, a dangerous vehicle is extracted during the line changing. Third, the condition is determined to wm or not by calculating the moving direction, relative distance md relative velocity. To set up the ROI, the only one side lane is detected and the interested region is expanded. Using the coordinate transformation method, the accuracy of the ROI raised. To correctly extract the vehicle on the neighboring lane, the Adaptive Background Update method and Image Segmentation method which uses the feature of the travelling road are used. The object which is extracted by the dangerous vehicle is calculated the relative distance, the relative velocity and the moving average. And then in order to ring, the direction of the vehicle and the condition for safe and unsafe is determined. As minimizes the interested region and uses the feature of the travelling road, the computational quantity is reduced and the accuracy is raised and a stable result on a travelling road images which demands a high speed calculation is showed.

• Journal of Platform Technology
• /
• v.9 no.2
• /
• pp.46-54
• /
• 2021

Recently, interest in the safety of Self-driving has been increasing. Self-driving have been studied and developed by many universities, research centers, car companies, and companies of other industries around the world since the middle 1980s. In this study, we propose the automatic extraction method of the threatening obstacle on the Road for the Self-driving. A threatening obstacle is defined in this study as a comparatively large object at center of the image. First of all, an input image and its decreased resolution images are segmented. Segmented areas are classified as the outer or the inner area. The outer area is adjacent to boundaries of the image and the other is not. Each area is merged with its neighbors when adjacent areas are included by a same area in the decreased resolution image. The Obstacle area and Non Obstacle area are selected from the inner area and outer area respectively. Obstacle areas are the representative areas for the obstacle and are selected by using the information about the area size and location. The Obstacle area and Non Obstacle area consist of the threatening obstacle on the road. Through experiments, we expect that the proposed method will be able to reduce accidents and casualties in Self-driving.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.1
• /
• pp.55-63
• /
• 2021

The roughness of the road is an important factor directly connected to the ride comfort, and is an evaluation item for functional evaluation and pavement quality management of the road. In this study, data on the road surface were acquired using the latest 3D geospatial information construction technology of ground LiDAR, drone photogrammetry, and drone LiDAR, and the accuracy and roughness of each method were analyzed. As a result of the accuracy evaluation, the average accuracy of terrestrial LiDAR were 0.039m, 0.042m, 0.039m RMSE in X, Y, Z direction, and drone photogrammetry and drone LiDAR represent 0.072~0.076m, 0.060~0.068m RMSE, respectively. In addition, for the roughness analysis, the longitudinal and lateral slopes of the target section were extracted from the 3D geospatial information constructed by each method, and the design values were compared. As a result of roughness analysis, the ground LiDAR showed the same slope as the design value, and the drone photogrammetry and drone LiDAR showed a slight difference from the design value. Research is needed to improve the accuracy of drone photogrammetry and drone LiDAR in measurement fields such as road roughness analysis. If the usability through improved accuracy can be presented in the future, the time required for acquisition can be greatly reduced by utilizing drone photogrammetry and drone LiDAR, so it will be possible to improve related work efficiency.