AUTOMATIC ROAD EXTRACTION FROM 1M-RESOLUTION SATELLITE IMAGES

Lee, Hae Yeoun,Park, Won Kyu,Lee, Heung Kyu 대한원격탐사학회 2000 International Symposium on Remote Sensing Vol.16 No.1

This paper addresses the extraction of roads from 1m-resolution satellite images. Among v~u-ious applications of satellite images, it is important to extract roads for Geographic Information System and urban planning, ete. Many approaches have been studied for airborne and low-resolution satellite images. However, these methods cannot be directly applied to high-resolution satellite images mainly because images contain more noises and roads are not line but regions. In this paper, we propose a region-based strategy for the extraction of roads from 1m-resolution satellite images. Firstly, by using an intensity-based segmentation scheme, regions representing road primitives are efficiently segmented. Secondly, road candidates are selected by region adjacency graph based an a prior knowledge about roads. Then, by comparing the profiles of road candidates with its adjacent profiles, roads can be expanded from road candidates. Using this strategy, we can currently extract most of roads identifiable by human in 1 m-resolution IKONOS satellite images.

Imaging Device Identification using Sensor Pattern Noise Based on Wiener Filtering

Hae-Yeoun Lee(이해연) 대한전기학회 2016 전기학회논문지 Vol.65 No.12

Multimedia such as image, audio, and video is easy to create and distribute with the advance of IT. Since novice uses them for illegal purposes, multimedia forensics are required to protect contents and block illegal usage. This paper presents a multimedia forensic algorithm for video to identify the device used for acquiring unknown video files. First, the way to calculate a sensor pattern noise using Wiener filter (W-SPN) is presented, which comes from the imperfection of photon detectors against light. Then, the way to identify the device is explained after estimating W-SPNs from the reference device and the unknown video. For the experiment, 30 devices including DSLR, compact camera, smartphone, and camcorder are tested and analyzed quantitatively. Based on the results, the presented algorithm can achieve the 96.0% identification accuracy.

Satellite Image Processing Software for Value-Added Products

Lee, Hae-Yeoun,Park, Won-Kyu,Kim, Seung-Bum,Kim, Tae-Jung,Yoon, Tae-Hun,Shin, Dong-Seok,Lee, Heung-Kyu The Korean Society of Remote Sensing 1999 大韓遠隔探査學會誌 Vol.15 No.4

To extract value-added products which are important in scientific area and practical life, e.g. digital elevation models, ortho-rectified images and geometric corrected images, Satellite Technology Research Center at Korea Advanced Institute of Science and Technology has developed a satellite image processing software called "Valadd-Pro". In this paper, "Valadd-Pro" software is briefly introduced and its main components such as precise geometric correction, ortho-rectification and digital elevation model extraction component are described. The performance of "Valadd-Pro" software was assessed on 10m resolution 6000 $\times$ 6000 SPOT panchromatic images (60km $\times$ 60km) using ground control points from GPS measurements. The height accuracy was measured by comparing our results with 100m resolution $DTEDs^{1)}$ produced by USGS and 60m resolution DEMs generated from digitized contours produced by National Geography Institute. Also, to show the superior performance of "Valadd-Pro" software, we compared the performance with that of commonly used PCI$\circledR$ commercial software. Based on the results, the geometric correction of "Valadd-Pro" software needs fewer ground control points than that of PCI$\circledR$ software and the ortho-rectification of "Valadd-Pro" software shows similar performance to that of PCI$\circledR$ software. In the digital elevation model extraction, "Valadd-Pro" software is two times more accurate and four times faster than PCI$\circledR$ software.ccurate and four times faster than PCI$\circledR$ software.

Satellite Image Processing Software for Value-Added Products

Hae Yeoun Lee,Won Kyu Park,Seung Bum Kim,Tae Jung Kim,Dong Seok Shin,Heung Kyu Lee 大韓遠隔探査學會 1999 大韓遠隔探査學會誌 Vol.15 No.4

To extract value-added products which are important in scientific area and practical life, e.g. digital elevation models, ortho-rectified images and geometric corrected images, Satellite Technology Research Center at Korea Advanced Institute of Science and Technology has developed a satellite image processing software called "Valadd-Pro". In this paper, "Valadd-Pro" software is briefly introduced and its main components such as precise geometric correction, ortho-rectification and digital elevation model extraction component are described. The performance of "Valadd-Pro" software was assessed on 10m resolution 6000×6000 SPOT panchromatic images (60km×60km) using ground control points from GPS measurements. The height accuracy was measured by comparing our results with 100m resolution DTEDs produced by USGS and 60m resolution DEMs generated from digitized contours produced by National Geography Institute. Also, to show the superior performance of "Valadd-Pro" software, we compared the performance with that of commonly used PCI(R) commercial software. Based on the results, the geometric correction of "Valadd-Pro" software needs fewer ground control points than that of PCI(R) software and the ortho-rectification of "Valadd-Pro" software shows similar performance to that of PCI(R) software. In the digital elevation model extraction, "Valadd-Pro" software is two times more accurate and four times faster than PCI(R) software.

Screenshot identification by analysis of directional inequality of interlaced video

Lee, Ji-Won,Lee, Min-Jeong,Lee, Hae-Yeoun,Lee, Heung-Kyu Springer (Biomed Central Ltd.) 2012 EURASIP journal on image and video processing Vol.2012 No.-

Geometrically invariant watermarking: synchronization through circular Hough transform

Lee, Hae-Yeoun,Lee, Choong-hoon,Lee, Heung-Kyu Springer-Verlag 2007 MULTIMEDIA TOOLS AND APPLICATIONS Vol.34 No.3

Automatic Left Ventricle Segmentation Using Iterative Thresholding and an Active Contour Model With Adaptation on Short-Axis Cardiac MRI

Lee, Hae-Yeoun,Codella, Noel C. F.,Cham, Matthew D.,Weinsaft, Jonathan W.,Wang, Yi IEEE 2010 IEEE Transactions on Biomedical Engineering Vol.57 No.4

<P>An automatic left ventricle (LV) segmentation algorithm is presented for quantification of cardiac output and myocardial mass in clinical practice. The LV endocardium is first segmented using region growth with iterative thresholding by detecting the effusion into the surrounding myocardium and tissues. Then the epicardium is extracted using the active contour model guided by the endocardial border and the myocardial signal information estimated by iterative thresholding. This iterative thresholding and active contour model with adaptation (ITHACA) algorithm was compared to manual tracing used in clinical practice and the commercial MASS Analysis software (General Electric) in 38 patients, with Institutional Review Board (IRB) approval. The ITHACA algorithm provided substantial improvement over the MASS software in defining myocardial borders. The ITHACA algorithm agreed well with manual tracing with a mean difference of blood volume and myocardial mass being 2.9 ± 6.2 mL (mean ± standard deviation) and -0.9 ± 16.5 g, respectively. The difference was smaller than the difference between manual tracing and the MASS software (approximately -20.0 ± 6.9 mL and -1.0 ± 20.2 g, respectively). These experimental results support that the proposed ITHACA segmentation is accurate and useful for clinical practice.</P>

Long-term endocrine outcome of suprasellar arachnoid cysts

Lee, Ji Yeoun,Lee, Young Ah,Jung, Hae Woon,Chong, Sangjoon,Phi, Ji Hoon,Kim, Seung-Ki,Shin, Choong-Ho,Wang, Kyu-Chang Journal of Neurosurgery Publishing Group 2017 Journal of neurosurgery. Pediatrics Vol.19 No.6

<B>OBJECTIVE</B><P>Due to their distinct location, suprasellar arachnoid cysts are known to cause a wide variety of problems, such as hydrocephalus, endocrine symptoms, and visual abnormalities. The long-term outcome of these cysts has not been elucidated. To find out the long-term outcome of suprasellar arachnoid cysts, a retrospective review of the patients was performed. The neurological and endocrine symptoms were thoroughly reviewed.</P><B>METHODS</B><P>Forty-five patients with suprasellar arachnoid cysts, with an average follow-up duration of 9.7 years, were enrolled in the study. A comprehensive review was performed of the results of follow-up regarding not only neurological symptoms but also endocrine status. The outcomes of 8 patients who did not undergo operations and were asymptomatic or had symptoms unrelated to the cyst were included in the series.</P><B>RESULTS</B><P>Surgery was most effective for the symptoms related to hydrocephalus (improvement in 32 of 32), but endocrine symptoms persisted after surgery (4 of 4) and required further medical management. More surprisingly, a fairly large number of patients (14 of 40; 1 was excluded because no pre- or postoperative endocrine evaluation was available) who had not shown endocrine symptoms at the time of the initial diagnosis and treatment later developed endocrine abnormalities such as precocious puberty and growth hormone deficiency. The patients with endocrine symptoms detected during the follow-up included those in both the operated (n = 12 of 32) and nonoperated (n = 2 of 8) groups who had been stable during follow-up since the initial diagnosis.</P><B>CONCLUSIONS</B><P>This study implies that patients with suprasellar arachnoid cysts can develop late endocrine problems during follow-up, even if other symptoms related to the cyst have been successfully treated. Hence, patients with these cysts need long-term follow-up for not only neurological symptoms but also endocrine abnormalities.</P>

Deregulation of Retinaldehyde Dehydrogenase 2 Leads to Defective Angiogenic Function of Endothelial Colony–Forming Cells in Pediatric Moyamoya Disease

Lee, Ji Yeoun,Moon, Youn Joo,Lee, Hae-Ock,Park, Ae-Kyung,Choi, Seung-Ah,Wang, Kyu-Chang,Han, Jung Woo,Joung, Je-Gun,Kang, Hyun Seung,Kim, Jeong Eun,Phi, Ji Hoon,Park, Woong-Yang,Kim, Seung-Ki American Heart Association 2015 Arteriosclerosis, thrombosis, and vascular biology Vol.35 No.7

<P>Objective-- Moyamoya disease (MMD) is a common cause of childhood stroke, in which the abnormal function of the endothelial colony-forming cell (ECFC) plays a key role in the pathogenesis of the disease. This study was designed to identify genes involved in MMD pathogenesis using gene expression profiling and to understand the defective function of MMD ECFCs. Approach and Results-- We compared gene expression profiles of ECFCs isolated from patients with MMD and normal controls. Among the differentially expressed genes, we selected a gene with the most downregulated expression, retinaldehyde dehydrogenase 2 (RALDH2). The activity of RALDH2 in MMD ECFCs was assessed by in vitro tube formation assay and in vivo Matrigel plug assay in the presence of all-trans retinoic acid. The transcriptional control of RALDH2 was tested using ChIP assays on acetyl-histone H3. In the results, MMD ECFCs inefficiently formed capillary tubes in vitro and capillaries in vivo, a defect restored by all-trans retinoic acid treatment. Knockdown of RALDH2 mRNA in normal ECFCs also induced decreased activity of capillary formation in vitro. The decreased level of RALDH2 mRNA in MMD ECFCs was attributed to defective acetyl-histone H3 binding to the promoter region. Conclusions-- From these results, we conclude that the expression of RALDH2 was epigenetically suppressed in ECFCs from patients with MMD, which may play a key role in their functional impairment.</P>

Epipolarity Analysis for Linear Pushbroom Imagery

Lee, Hae Yeoun,Park, Won Kyu 대한원격탐사학회 2001 International Symposium on Remote Sensing Vol.17 No.1

This paper addresses the epipolar geometry of the linear pushbroom imagery. Two images of a single scene are related by epipolar geometry which contain all geometric information and is essential for the three dimensional reconstruction of the scene in computer vision and remote sensing. It is told that the epipolar geometry of the linear pushbroom sensor is different from that of the perspective one. In this paper, an accurate epipolarity model without the ephemeris data is proposed. Also, the various epipolarity models such as the one of the perspective and aerial imagery, the one based an the simplified pushbraom model and the one based on the reduced pushbroom model is reviewed and analyzed for linear pushbroom imagery. To analyze the accuracy of the proposed epipolarity model and others, we devised the quantitative accuracy measure for the epipalarity analysis and tested using two linear pushbroom sensors from SPOT and KOMPSAT satellite. Based on the results, the epipolarity model based on the reduced pushbroom model is accurate, but requires to the ephemeris data. However, the proposed epipolarity model is considerably accurate although the ephemeris data is not used. This model is very useful when the ephemeris data cannot be acquired such as IKONOS energies are not accurate.