• The Journal of the Korea Contents Association
• /
• v.19 no.1
• /
• pp.125-131
• /
• 2019

A vanishing point is a point on an image to which parallel lines projected from a real space gather. A vanishing point in a road space provides important spatial information. It is possible to improve the position of an extracted lane or generate a depth map image using a vanishing point in the road space. In this paper, we propose a method of detecting vanishing points on images taken from a vehicle's point of view using Deep Neural Network (DNN) and Histogram of Oriented Gradient (HoG). The proposed algorithm is divided into a HoG feature extraction step, in which the edge direction is extracted by dividing an image into blocks, a DNN learning step, and a test step. In the learning stage, learning is performed using 2,300 road images taken from a vehicle's point of views. In the test phase, the efficiency of the proposed algorithm using the Normalized Euclidean Distance (NormDist) method is measured.

• Nuclear Engineering and Technology
• /
• v.54 no.1
• /
• pp.49-60
• /
• 2022

Motivated by the high-resolution properties of high-order Weighted Essentially Non-Oscillatory (WENO) and flux limiter (FL) for steep-gradient problems and the robust convergence of Jacobian-free Newton-Krylov (JFNK) methods for nonlinear systems, the preconditioned JFNK fully implicit high-order WENO and FL schemes are proposed to solve the transient two-phase two-fluid models. Specially, the second-order fully-implicit BDF2 is used for the temporal operator and then the third-order WENO schemes and various flux limiters can be adopted to discrete the spatial operator. For the sake of the generalization of the finite-difference-based preconditioning acceleration methods and the excellent convergence to solve the complicated and various operational conditions, the random vector instead of the initial condition is skillfully chosen as the solving variables to obtain better sparsity pattern or more positions of non-zero elements in this paper. Finally, the WENO_JFNK and FL_JFNK codes are developed and then the two-phase steep-gradient problem, phase appearance/disappearance problem, U-tube problem and linear advection problem are tested to analyze the convergence, computational cost and efficiency in detailed. Numerical results show that WENO_JFNK and FL_JFNK can significantly reduce numerical diffusion and obtain better solutions than traditional methods. WENO_JFNK gives more stable and accurate solutions than FL_JFNK for the test problems and the proposed finite-difference-based preconditioning acceleration methods based on the random vector can significantly improve the convergence speed and efficiency.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_3
• /
• pp.891-907
• /
• 2023

Compact Advanced Satellite 500-4 (CAS500-4) is scheduled to be launched to collect high spatial resolution data focusing on vegetation applications. To achieve this goal, accurate surface reflectance retrieval through atmospheric correction is crucial. Therefore, a machine learning-based atmospheric correction algorithm was developed to simulate atmospheric correction from a radiative transfer model using Sentinel-2 data that have similarspectral characteristics as CAS500-4. The algorithm was then evaluated mainly for forest areas. Utilizing the atmospheric correction parameters extracted from Sentinel-2 and GEOKOMPSAT-2A (GK-2A), the atmospheric correction algorithm was developed based on Random Forest and Light Gradient Boosting Machine (LGBM). Between the two machine learning techniques, LGBM performed better when considering both accuracy and efficiency. Except for one station, the results had a correlation coefficient of more than 0.91 and well-reflected temporal variations of the Normalized Difference Vegetation Index (i.e., vegetation phenology). GK-2A provides Aerosol Optical Depth (AOD) and water vapor, which are essential parameters for atmospheric correction, but additional processing should be required in the future to mitigate the problem caused by their many missing values. This study provided the basis for the atmospheric correction of CAS500-4 by developing a machine learning-based atmospheric correction simulation algorithm.

• Korean Journal of Ecology and Environment
• /
• v.37 no.2 s.107
• /
• pp.193-204
• /
• 2004

This study was conducted to evaluate both temporal and spatial dynamics of phytoplankton community and environmental parameters in a newly made reclaimed esturine lake (L. Hwaong). Monthly sampling was conducted at 4 sites covering the longest transect of longitudinal gradient of the lake from June to November, 2002. Total 5 classes 8 orders 26 families 83 genus 192 species were identified at all study sites during the study period. Phytoplankton total cell density ranged 24${\sim}$ 1,882 cells $mL^{-1}$ and highly varied both temporally and spatially. Total cell density was significantly related with salinity, pH, BOD, COD, SS, TN and TP concentration. Diatom density also was significantly correlated with salinity, SS, BOD, COD and TN concentration, Although there was spatial difference, a longtudinal gradient appeared in phytoplankton cell density, Chl-a, TN and TP concentration from the mouth of river in June and August. In conclusion, phytoplakton community structure was dominated by diatoms (Bacillariophyceae), and appeared to be largely influenced by salinity, precipitation, and nutrients during the summer and the fall.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1329-1341
• /
• 2022

Destroying the marine environment and marine ecosystem and causing marine accidents, marine debris is generated every year, and among them, submerged marine debris is difficult to identify and collect because it is on the seabed. Therefore, deep-learning-based semantic segmentation was experimented on waste fish nets and waste ropes using underwater images to identify efficient collection and distribution. For segmentation, a high-resolution network (HRNet), a state-of-the-art deep learning technique, was used, and the performance of each optimizer was compared. In the segmentation result fish net, F1 score=(86.46%, 86.20%, 85.29%), IoU=(76.15%, 75.74%, 74.36%), For the rope F1 score=(80.49%, 80.48%, 77.86%), IoU=(67.35%, 67.33%, 63.75%) in the order of adaptive moment estimation (Adam), Momentum, and stochastic gradient descent (SGD). Adam's results were the highest in both fish net and rope. Through the research results, the evaluation of segmentation performance for each optimizer and the possibility of segmentation of marine debris in the latest deep learning technique were confirmed. Accordingly, it is judged that by applying the latest deep learning technique to the identification of submerged marine debris through underwater images, it will be helpful in estimating the distribution of marine sedimentation debris through more accurate and efficient identification than identification through the naked eye.

• Journal of the Korean association of regional geographers
• /
• v.9 no.4
• /
• pp.481-492
• /
• 2003

This study aims to analyze quantitatively changes of forest and cropland landscape due to cropland increase toward higher mountain slope during 9 years from 1993 to 2002, using Landsat images and field survey in the vicinity of Manpo and Gangae cities, Jagang Province, Northwest Korea, During 9 years, cropland has increased as 49.9%, forest area decreased as 16%. The spatial characteristics of cropland changes present that average elevation of cropland are increased from 381m of 1993 year to 412m of 2002 year, and average gradient increased from $10^{\circ}$ to $13^{\circ}$. In increased area of cropland during 9 years, the average elevation is 455m, and average gradient is $15^{\circ}$ with maximum gradient $70^{\circ}$. Analysis of the patch phenomena by fragmentation of vegetation landscape show that number of patch increased from 394 to 1,241 and also values of shape index, and fractal dimension of vegetation are increased slightly, during 9 years. Croplands have developed mainly in mountain slopes of elevation between 450 to 750m. For improvement of agricultural productivity, it should be required political and social stabilization, international and South Korea's assistance, and restoration of mountain forest.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.38 no.1
• /
• pp.104-115
• /
• 2001

A new content-based color image retrieval method integrating the features of the color adjacency and the gradient is proposed in this paper. As the most used feature of color image, color histogram has its own advantages that it is invariant to the changes in viewpoint and the rotation of the image etc., and the computation of the feature is simple and fast. However, it is difficult to distinguish those different images having similar color distributions using histogram-based image retrieval, because the color histogram is generated on uniformly quantized colors and the histogram itself contains no spatial information. And another shortcoming of the histogram-based image retrieval is the storage of the features is usually very large. In order to prevent the above drawbacks, the gradient that is the largest color difference of neighboring pixels is calculated in the proposed method instead of the uniform quantization which is commonly used at most histogram-based methods. And the color adjacency information which indicates major color composition feature of an image is extracted and represented as a binary form to reduce the amount of feature storage. The two features are integrated to allow the retrieval more robust to the changes of various external conditions.

• Journal of Korean Society for Geospatial Information Science
• /
• v.11 no.1 s.24
• /
• pp.21-28
• /
• 2003

This study was aimed at suggest a proper planning method to select a optimal forest road network in mountains forest using GIS(Geographic Information System). To examine the field applicability, the method was applied to the National Forest at Mt. Kumsan in Namhae-gun, Gyungsangnam-do. The main results from altogether six alternative road route plans were derived from these criteria obtained the alternative route plan No.2 has two layout criteria, longitudinal gradient and earth work volume, and it showed similar pattern of existing forest road network which was designed mainly ground slope and longitudinal gradient. The alternative route plan No.6 has four criteria, longitudinal gradient, earth work volume, investment effect and landscape impact. It was different for the lowest forest road density among the alternatives and the pattern of the forest road layout was radial form, which was also quite different to other alternatives. For optimal forest road network planning, GIS provide the efficient and resonable solutions for decision making to provide the support for evaluation about various alternative road networks. If detailed inventory and relevant data are provided and also clear and objective indicators for evaluations are set up, it could be applied to preliminary analysis and detail planning stage to prevent undesirable effect such the land slide and soil erosion due to inadequate planning for forest road network.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.6
• /
• pp.1201-1208
• /
• 2020

A mass of seawater with similar properties in the ocean is called a water mass, and the front is a sea area where two masses of different properties meet. The gradient algorithm is a method of extracting where the sea water temperature pixel changes rapidly assuming that the slope is large, and the place with the large slope is assumed to be a front. This method is able to process large amounts of satellite data at once. Therefore, in this study, we tried to find the front lines in the sea area around the Korean Peninsula by using a gradient algorithm. The study data used gridded sea surface temperature satellite data. The resolution was 1/4°, and the monthly average data from January 1993 to December 2018 were used. There were major five fronts representatively, China Coastal Front, South Sea Coastal Front, Kuroshio Front/ Kuroshio Extension Front, Subpolar Front and the Subarctic Front. As a result of comparing the distribution of front by season, more types of front were distributed in winter and spring than in summer and autumn, and the distribution range was wider.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.3D
• /
• pp.445-450
• /
• 2008

Recently, environmental and ecological approaches is in progress in urban stream, especially the guarantee of instream flow becomes very important. In this paper, it is suggested that water volume estimation method utilizing the topographical shape data obtained by field investigation and satellite image to manage the urban stream efficiently. The data obtained at Gap River is the study area are analysed and those results are as belows. First, surveying to investigate topographic shape characteristics of urban stream is carried out. In details, the gradient characteristics from water surface to bottom in case of sand area and in case of grass area are 0.013 and 0.065 respectively. In conclusion, the gradient characteristic of grass area is five times bigger than that of sand area. Besides, IKONOS image is classified by spectrum analysis and Minimum Distance Method and the sand area extraction method by the generalization method as Median filter is suggested to calculate water volume. Finally, mapping process on the sand area extracted from the topographical shape field data in river and satellite images is carried out by the GIS spatial analysis. And on the assumption that the water level was 1m at that time when satellite image was taken, the water volume was $225,258m^3$. It is clarified that the effect of water volume improvement was about 10.5% in comparison with water volume that had no consideration on the gradient characteristics of sand-bar.