• 전자공학회논문지
• /
• 제50권4호
• /
• pp.137-143
• /
• 2013

고해상도 위성영상은 기상관측, 지형관측, 원격탐사, 군사시설감시, 문화재보호 등 많은 분야에서 이용된다. 위성영상은 동일한 위성영상 시스템에서 획득한 영상이라 할지라도 하드웨어(광학장치, 위성의 운용고도, 영상 센서 등)의 조건에 따라서 해상도가 저하된 영상들이 발생한다. 따라서 위성이 발사된 이후에는 이러한 해상도가 저하된 영상들의 해상도 향상을 위해서 영상시스템의 하드웨어를 변경하는 것은 불가능하므로 위성영상 자체를 이용하여 해상도를 향상시키는 방법이 필요하다. 본 논문에서는 이러한 저해상도 위성영상을 이용하여 해상도를 향상시키는 방법으로 SR(Super Resolution) 알고리즘을 사용하였다. SR 알고리즘은 다수의 저해상도 영상들의 정합을 통해 영상의 해상도를 향상시키는 알고리즘이다. 하지만 위성영상에서는 동일 지역에 대한 여러 장의 영상을 획득하기 어렵다. 따라서 본 논문에서는 이러한 문제점을 해결하기 위해 어파인 변환(Affine Transform)및 투영 변환(Projection Transform)을 적용 후 영상에 대한 기하학적 변화를 보정하여 SR 알고리즘을 수행하였다. 그 결과 SR 알고리즘만 적용한 영상보다 어파인 변환과 투영 변환을 거친 후 SR 알고리즘을 적용한 영상에서 해상도가 확실하게 더 증가되는 것을 확인하였다.

• 인터넷정보학회논문지
• /
• 제10권4호
• /
• pp.13-23
• /
• 2009

본 논문에서 우리는 위성 영상에 대하여 객체를 지역화한 접근을 제안한다. 우리의 방법은 서술 벡터에 기반한 특징 정합 방법이다. 객체를 지역화하는 방법은 SIFT(Scale Invariant Feature Transform)를 적용시킨다. 먼저, 위성영상의 키포인트를 찾고, 키포인트의 서술 벡터를 일반화한다. 그리고 서술 벡터간에 유사성을 측정하여 키포인트를 매칭시킨다. 마지막으로, 키포인트의 인접 픽셀값에 가중치를 주어 객체에서 위치를 결정한다. SIFT를 이용한 이 실험은 다양한 스케일과 어파인 변환에 대해 좋은 결과를 산출하였다. 본 논문에서 제안된 방법은 구글 어스의 위성영상을 사용하였다.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 1998년도 Proceedings of International Symposium on Remote Sensing
• /
• pp.252-256
• /
• 1998

Accurate mapping of satellite images is one of the most important Parts in many remote sensing applications. Since the position and the attitude of a satellite during image acquisition cannot be determined accurately enough, it is normal to have several hundred meters' ground-mapping errors in the systematically corrected images. The users which require a pixel-level or a sub-pixel level mapping accuracy for high-resolution satellite images must use a number of Ground Control Points (GCPs). In this paper, the performance of two geometric correction algorithms is tested and compared. One is the polynomial warping algorithm which is simple and popular enough to be implemented in most of the commercial satellite image processing software. The other is full camera modelling algorithm using Physical orbit-sensor-Earth geometry which is used in satellite image data receiving, pre-processing and distribution stations. Several criteria were considered for the performance analysis : ultimate correction accuracy, GCP representatibility, number of GCPs required, convergence speed, sensitiveness to inaccurate GCPs, usefulness of the correction results. This paper focuses on the usefulness of the precision correction algorithm for regular image pre-processing operations. This means that not only final correction accuracy but also the number of GCPs and their spatial distribution required for an image correction are important factors. Both correction algorithms were implemented and will be used for the precision correction of KITSAT-3 images.

• 대한원격탐사학회지
• /
• 제39권4호
• /
• pp.395-407
• /
• 2023

Super-resolution (SR) has great significance in image processing because it enables downstream vision tasks with high spatial resolution. Recently, SR studies have adopted deep learning networks and achieved remarkable SR performance compared to conventional example-based methods. Deep-learning-based SR models generally require low-resolution (LR) images and the corresponding high-resolution (HR) images as training dataset. Due to the difficulties in obtaining real-world LR-HR datasets, most SR models have used only HR images and generated LR images with predefined degradation such as bicubic downsampling. However, SR models trained on simple image degradation do not reflect the properties of the images and often result in deteriorated SR qualities when applied to real-world images. In this study, we propose an image degradation model for HR satellite images based on the modulation transfer function (MTF) of an imaging sensor. Because the proposed method determines the image degradation based on the sensor properties, it is more suitable for training SR models on remote sensing images. Experimental results on HR satellite image datasets demonstrated the effectiveness of applying MTF-based filters to construct a more realistic LR-HR training dataset.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 1999년도 Proceedings of International Symposium on Remote Sensing
• /
• pp.37-42
• /
• 1999

A compact imaging system, the Multi-spectral Earth Imaging System (MEIS) was developed and operated on an engineering test satellite, KITSAT-3 at the orbital altitude of 720 km. The MEIS takes multi-spectral images of the earth's surface with the swath width of 48 km and the ground sampling distance of 13.8 m in three spectral bands. A brief technical description of the KITSAT-3 MEIS and the result from its initial operation since early June, 1999 are presented. The quality of images produced by the KITSAT-3 MEIS was found comparable to that of images from existing commercial earth observation satellites from its preliminary assessment.

• Korean Journal of Geomatics
• /
• 제3권1호
• /
• pp.53-57
• /
• 2003

As the Korea Multi-Purpose Satellite-I (KOMPSAT-1) satellite can roll tilt up to $\pm$45$^{\circ}$, we have analyzed some KOMPSAT-1 EOC images taken at different tilt angles for this study. The required ground coordinates for bundle adjustment and geometric accuracy are obtained from the digital map produced by the National Geography Institution, at a scale of 1:5,000. Followings are the steps taken for the tilting angle of KOMPSAT-1 to be present in the evaluation of geometric accuracy of each different stereo image data: Firstly, as the tilting angle is different in each image, the characteristic of satellite dynamic must be determined by the sensor modeling. Then the best sensor modeling equation should be determined. The result of this research, the difference between the RMSE values of individual stereo images is mainly due to quality of image and ground coordinates instead of tilt angle. The bundle adjustment using three KOMPSAT-1 stereo pairs, first degree of polynomials for modeling the satellite position, were sufficient.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1998년도 하계학술대회 논문집 G
• /
• pp.2209-2211
• /
• 1998

This paper presents the matching algorithm with the satellite images using the image warping method. Two stereo images, which are used for the DEM(Digital Elevation Model) extraction, are generally distorted because the images are acquired at different locations and angles. Therefore, the matching Process can't be executed with the original images. To solve this problem, a dynamic triangular image warping method is proposed. At first, the initial matching is executed with seed point, and then, using the matched points from the initial matching, the distorted images is compensated. We experimented this algorithm with the parts of the $6000{\times}6000$ SPOT satellite images. The experiment results show this algorithm is superior to other warping algorithm.

• 대한원격탐사학회지
• /
• 제16권4호
• /
• pp.347-353
• /
• 2000

The purpose of this paper is to extract DEM (Digital Elevation Model) using KOMPSAT images. DEM extraction consists of three parts. First part is the modeling of satellite position and attitude, second part is the matching of two images to find corresponding points of them and third part is to calculate the elevation of each point by using the result of the first and second part. The position and attitude modeling of satellite is processed by using GCPs. Area based matching method is used to find the corresponding points between the stereo satellite images. The elevation of each point is calculated using the exterior orientation information obtained from sensor modeling and the disparity from the stereo matching. In experiment, the KOMPSAT images, 2592$\times$2796 panchromatic images are used to extract DEM. The experiment result show the DEM using KOMPSAT images.

• 대한원격탐사학회지
• /
• 제38권6_4호
• /
• pp.1901-1910
• /
• 2022

뉴스페이스(new space) 시대가 도래함에 따라 국내 KOMPSAT-3·3A 위성영상과 해외 위성영상과의 글로벌 융합활용 기술확보가 대두되고 있다. 일반적으로 다중센서 위성영상은 취득 당시의 다양한 외부요소로 인해 영상 간 상대적인 기하오차(relative geometric error)가 발생하며, 이로 인해 위성영상 산출물의 품질이 저하된다. 따라서 본 연구에서는 KOMPSAT-3·3A 위성영상과 해외 위성영상 간 존재하는 상대기하오차를 최소화하기 위한 정밀영상정합(fine-image registration) 방법론을 제안한다. KOMPSAT-3·3A 위성영상과 해외 위성영상 간 중첩영역을 선정한 후 두 영상 간 공간해상도를 통일한다. 이어서, 특징 및 영역 기반 정합기법을 결합한 형태의 하이브리드(hybrid) 정합기법을 이용하여 정합점(tie-point)을 추출한다. 그리고 피라미드(pyramid) 영상 기반의 반복적 정합을 수행하여 정밀영상정합을 수행한다. KOMPSAT-3·3A 위성영상과 Sentinel-2A 및 PlanetScope 영상을 이용하여 제안기법의 정확도 및 성능을 평가하였다. 그 결과, Sentienl-2A 영상 기준 평균 Root Mean Square Error (RMSE) 1.2 pixels, PlanetScope 영상 기준 평균 RMSE 3.59 pixels의 정확도가 도출되었다. 이를 통해 제안기법을 이용하여 효과적으로 정밀영상정합을 수행할 수 있을 것으로 사료된다.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2005년도 Proceedings of ISRS 2005
• /
• pp.47-51
• /
• 2005

Using satellite images, an optimal solution to water motion has been presented in this study. Since temperature patterns are suitable tracers in water motion, Sea Surface Temperature (SST) images of Caspian Sea taken by MODIS sensor on board Terra satellite have been used in this study. Two daily SST images with 24 hours time interval are used as input data. Computation of templates correspondence between pairs of images is crucial within motion algorithms using non-rigid body objects. Image matching methods have been applied to estimate water body motion within the two SST images. The least squares matching technique, as a flexible technique for most data matching problems, offers an optimal spatial solution for the motion estimation. The algorithm allows for simultaneous local radiometric correction and local geometrical image orientation estimation. Actually, the correspondence between the two image templates is modeled both geometrically and radiometrically. Geometric component of the model includes six geometric transformation parameters and radiometric component of the model includes two radiometric transformation parameters. Using the algorithm, the parameters are automatically corrected, optimized and assessed iteratively by the least squares algorithm. The method used in this study, has presented more efficient and robust solution compared to the traditional motion estimation schemes.