• Journal of the Korea Computer Graphics Society
• /
• v.13 no.3
• /
• pp.7-10
• /
• 2007

We propose an image filtering technique that removes various image styles. To destylize a given image, we define image styles as repeated patterns existing in the image. For dll pixels of the image, we compute image styles as style vectors. We remove image styles by using bilateral filtering based on these style vectors. Destylization results show well smoothed images while preserving feature boundaries. Our method effectively removes image styles and reveals image structures clearly, and results can be applied to several applications such as texture transfer.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10c
• /
• pp.580-582
• /
• 1998

인터넷의 보급으로 인하여 디지털 데이터의 복제가 확산됨에 따라 멀티미디어 데이터에 대한 소유권 보호와 인증에 대한 문제가 대두되고 있는 실정이다. 본 논문에서는 디지털 영상을 다중해상도 표현이 가능한 웨이브릿 변환(Wavelet Transform)을 통하여 저주파수 영역에 인간 시각으로 지각 할 수 없는 워터마크를 삽입하고, 삽입된 워터마크의 영상을 인증하기 위한 방법으로 오류 역전파 학습 알고리즘(Error Backpropagation)을 이용한 신경회로망적 접근방법을 제안한다. 워터마크를 추출하기 위해서는 원영상이 필요하고, 내장된 워터마크가 손실 압축과 필터링 등의 일반적인 영상 처리에 강인함을 실험 결과를 증명하고, 제안한 신경회로망적 접근방법이 좋은 결과를 나타냄으로 실험을 통하여 증명하였다.

• Progress in Medical Physics
• /
• v.14 no.3
• /
• pp.167-174
• /
• 2003

In the radiation treatment planning (RTP) process, especially for stereotactic radiosurgery (SRS), knowing the exact volume and shape and the precise position of a lesion is very important. Sometimes X-ray projection images, such as angiograms, become the best choice for lesion identification. However, while the exact target position can be acquired by bi-projection images, 3D target reconstruction from bi-projection images is considered to be impossible. The aim of this study was to reconstruct the 3D target volume from multiple projection images. It was assumed that we knew the exact target position in advance, and all processes were performed in Target Coordinates, where the origin was the center of the target. We used six projections: two projections were used to make a Reconstruction Box and four projections were for image acquisition. The Reconstruction Box was made up of voxels of 3D matrices. Projection images were transformed into 3D in this virtual box using a geometric back-projection method. The resolution and the accuracy of the reconstructed target volume were dependent on the target size. An algorithm was applied to an ellipsoid model and a horseshoe-shaped model. Projection images were created geometrically using C program language, and reconstruction was also performed using C program language and Matlab ver. 6(The Mathwork Inc., USA). For the ellipsoid model, the reconstructed volume was slightly overestimated, but the target shape and position proved to be correct. For the horseshoe-shaped model, reconstructed volume was somewhat different from the original target model, but there was a considerable improvement in determining the target volume.

• Journal of Korean Society for Geospatial Information Science
• /
• v.12 no.4 s.31
• /
• pp.3-12
• /
• 2004

It is necessary to manage water resources in rural areas in order to achieve proper development of new water resources, sustainable usage and reasonable distribution. This paper aims to analyze multi-temporal Landsat-7 ETM+data for soil moisture that is essential for crops in Ahnsung area. The ETM data was also fused with KOMPSAT-1 images in order to be used as backdrop watershed maps at first. Multi-temporal Images showed also the characteristics of soil moisture distribution. Images taken in April showed that rice paddy had as low reflectance as artificial features. Compared with April scenes, those taken in Hay and June showed wetness index increased in the rice paddies. The mountainous areas have almost constant moisture index, so the difference between the dates was very low while reservoirs and livers had dramatic changes. We can calculate total potential areas of distribution of moisture content within the basin and estimate the areas being sensitive to drought. Finally we can point out the sites of small rice paddies lack of water and visualize their distribution within the same basin. It can be said that multi-temporal Landsat-7 ETM+ and KOMPSAT data can be used to show broad drought with quick and simple analysis. Drought sensitiveness maps may enable the decision makers on rural water to evaluate the risk of drought and to measure mitigation, accompanied with proper data on the hydrological and climatic drought.

• 한국지형공간정보학회:학술대회논문집
• /
• 2002.03a
• /
• pp.150-155
• /
• 2002

SAR 시스템은 능동적 센서로 마이크로파라 불리우는 전자기파를 직접 지상에 보내고 돌아오는 신호의 위상과 진폭을 이용하여 영상으로 나타내는 간섭성 시스템이다. 이러한 영상의 특성으로 인해 날씨나 태양의 유 무에 상관없이 영상을 취득할 수 있는 장점이 있다. 또한, 최근에는 기존의 다중분광 위성영상과의 SAR 영상의 Data Fusion을 통해 지상의 새로운 정성적 정보를 취득하려는 시도 등 나날이 그 활용성이 증대되고 있는 상황이다. 그러나 SAR 영상의 광범위한 활용을 위해서는 먼저 영상의 지형보정이 선행되어야 한다. 따라서 본 연구에서는 SAR 영상의 활용을 위해서 선행되어야 할 지형보정의 알고리즘을 컴포넌트 기반의 프로그램으로 구현하고 대상연구지역에 대한 적용을 통해 그 활용성과 가능성을 보여주고자 한다. 연구대상지역은 ERS-1, ERS-2 SAR로 촬영된 대전광역시와 그 주변지역으로 해당 SAR 영상에 대하여 엄밀지형보정 알고리즘과 경사거리 영상을 지상거리 영상으로 변환하는 알고리즘을 개발하여 적용하였다. 실험결과 공칭해상도 30m의 ERS 영상에 대하여 39.7m(X방향으로 24.5m, Y방향으로 31.3m)의 수평오차를 나타내었으며 경사거리 영상의 지상거리 영상으로의 변환도 원활하게 수행됨을 알 수 있었다. 마지막으로 본 연구를 통해 연구된 모든 알고리즘은 컴포넌트 기반으로 설계하고 구현되어 향후 국내 SAR 처리기술 개발에 있어서 공유할 수 있도록 하였다.

• Journal of Cadastre & Land InformatiX
• /
• v.46 no.2
• /
• pp.239-251
• /
• 2016

Hyperspectral image is widely used for land cover classification because it has a number of narrow bands and allow each pixel to include much more information in comparison with previous multi-spectral image. However, Higher spectral resolution of hyperspectral image results in an increase in data volumes and a decrease in noise efficiency. SAM(Spectral Angle Mapping), a method based on vector inner product to compare spectrum distribution, is a highly valuable and popular way to analyze continuous spectrum of hyperspectral image. SAM is shown to be less accurate when it is used to analyze hyperspectral image for land cover classification using spectral library. this inaccuracy is due to the effects of atmosphere. We suggest a decision tree based method to compensate the defect and show that the method improved accuracy of land cover classification.

• The Journal of the Korea Contents Association
• /
• v.4 no.1
• /
• pp.18-23
• /
• 2004

In this paper, we propose a new visual feature extraction method for content-based image retrieval(CBIR) based on wavelet transform which has both spatial-frequency characteristic and multi-resolution characteristic. We extract visual features for each frequency band in wavelet transformation and use them to CBIR. The lowest frequency band involves spacial information of original image. We extract L feature vectors using fuzzy homogeneity in the wavelet domain, which consider both the wavelet coefficients and the spacial information of each coefficient. Also, we extract 3 feature vectors wing the energy values of high frequency bands, and store those to image database. As a query, we retrieve the most similar image from image database according to the 10 largest homograms(normalized fuzzy homogeneity vectors) and 3 energy values. Simulation results show that the proposed method has good accuracy in image retrieval using 90 texture images.

• Journal of Intelligence and Information Systems
• /
• v.29 no.4
• /
• pp.257-269
• /
• 2023

There is currently a burgeoning demand for image synthesis from photos and videos using deep learning models. Existing video synthesis models solely extract motion information from the provided video to generate animation effects on photos. However, these synthesis models encounter challenges in achieving accurate lip synchronization with the audio and maintaining the image quality of the synthesized output. To tackle these issues, this paper introduces a novel framework based on an image animation approach. Within this framework, upon receiving a photo, a video, and audio input, it produces an output that not only retains the unique characteristics of the individuals in the photo but also synchronizes their movements with the provided video, achieving lip synchronization with the audio. Furthermore, a super-resolution model is employed to enhance the quality and resolution of the synthesized output.

• Journal of Internet Computing and Services
• /
• v.21 no.4
• /
• pp.117-126
• /
• 2020

A maritime object detection system is an intelligent assistance system to maritime autonomous surface ship(MASS). It detects automatically floating debris, which has a clash risk with objects in the surrounding water and used to be checked by a captain with a naked eye, at a similar level of accuracy to the human check method. It is used to detect objects around a ship. In the past, they were detected with information gathered from radars or sonar devices. With the development of artificial intelligence technology, intelligent CCTV installed in a ship are used to detect various types of floating debris on the course of sailing. If the speed of processing video data slows down due to the various requirements and complexity of MASS, however, there is no guarantee for safety as well as smooth service support. Trying to solve this issue, this study conducted research on the minimization of computation volumes for video data and the increased speed of data processing to detect maritime objects. Unlike previous studies that used the Hough transform algorithm to find the horizon and secure the areas of interest for the concerned objects, the present study proposed a new method of optimizing a binarization algorithm and finding areas whose locations were similar to actual objects in order to improve the speed. A maritime object detection system was materialized based on deep learning CNN to demonstrate the usefulness of the proposed method and assess the performance of the algorithm. The proposed algorithm performed at a speed that was 4 times faster than the old method while keeping the detection accuracy of the old method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.4
• /
• pp.543-548
• /
• 2021

The higher the resolution of the image, the higher the satisfaction of the viewers of the image, and the super-resolution imaging has a considerable increase in research value among the fields of computer vision and image processing. In this study, the main features of low-resolution image LR are extracted mainly using deep learning super-resolution models. It learns and reconstructs the extracted features, and focuses on reconstruction-based algorithms that generate high-resolution image HR. In this paper, we investigate SRCNN and VDSR in a super-resolution algorithm model based on reconstruction. The structure and algorithm process of the SRCNN and VDSR model are briefly introduced, and the multi-channel and special form are also examined in the improved performance evaluation function, and understand the performance of each algorithm through experiments. In the experiment, an experiment was performed to compare the results of the SRCNN and VDSR models with the peak signal-to-noise ratio and image structure similarity, so that the results can be easily judged.