• 제어로봇시스템학회논문지
• /
• 제15권8호
• /
• pp.778-785
• /
• 2009

This paper presents a Gabor texture feature extraction method for classification of discolored Metal pad images using GPU(Graphics Processing Unit). The proposed algorithm extracts the texture information using Gabor filters and constructs a pattern map using the extracted information. Finally, the golden pad images are classified by utilizing the feature vectors which are extracted from the constructed pattern map. In order to evaluate the performance of the Gabor texture feature extraction algorithm based on GPU, a sequential processing and parallel processing using OpenMP in CPU of this algorithm were adopted. Also, the proposed algorithm was implemented by using Global memory and Shared memory in GPU. The experimental results were demonstrated that the method using Shared memory in GPU provides the best performance. For evaluating the effectiveness of extracted Gabor texture features, an experimental validation has been conducted on a database of 20 Metal pad images and the experiment has shown no mis-classification.

• 한국컴퓨터정보학회논문지
• /
• 제12권3호
• /
• pp.105-113
• /
• 2007

본 논문에서는 경계 중요도 맵과 영역 병합에 의한 영상 분할 방법을 제안한다. 경계 중요도 맵은 텍스쳐 경계 강도와 칼라 경계 강도의 조합에 의해 생성한다. 텍스쳐 경계 강도는 가버 필터 뱅크를 사용하여 다중 스케일과 방향에 따른 필터링 결과를 병합하여 생성하며 칼라 경계 강도는 HSI 칼라 모델의 H 성분에 대해 계산한다. 경계 중요 맵 영상에 대해서는 Watershed 변환을 통해 사전 영상 분할을 수행한다. Watershed 변환에 의한 영상 분할은 영역들이 과잉 분할되는 현상이 나타나므로 이를 개선하여 최종 영상 분할 결과를 생성한다. 이를 위해 우선 모폴로지 연산을 사용하여 경계 중요도 맵 영상에 대한 컨트라스트 향상과 마커 영역을 생성한다. 모폴로지 연산으로 과잉 분할 영역은 줄어들지만 여전히 상당수 존재하게 되므로 이를 극복하기 위해 영역 병합 과정을 수행한다. 영역 병합 단계에서는 영역 내부의 평균 칼라 및 가버 텍스쳐 벡터를 함께 사용함으로써 효과적으로 과잉 분할된 영역을 병합할 수 있도록 하였다. 제안한 방법은 다양한 자연 영상에 대해 실험하였으며 기존 방법과 결과를 비교하여 성능의 우수성을 확인하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권3호
• /
• pp.1121-1139
• /
• 2015

In texture-plus-depth format, depth map compression is an important task. Different from normal texture images, depth maps have less texture information, while contain many homogeneous regions separated by sharp edges. This feature will be employed to form an efficient depth map coding scheme in this paper. Firstly, the histogram of the depth map will be analyzed to find an appropriate threshold that segments the depth map into the foreground and background regions, allowing the edge between these two kinds of regions to be obtained. Secondly, the two regions will be encoded through rate distortion optimization with a shape adaptive wavelet transform, while the edges are lossless encoded with JBIG2. Finally, a depth-updating algorithm based on the threshold and the depth range is applied to enhance the quality of the decoded depth maps. Experimental results demonstrate the effective performance on both the depth map quality and the synthesized view quality.

• 융복합기술연구소 논문집
• /
• 제12권1호
• /
• pp.1-5
• /
• 2022

This paper describes an algorithm for generating a terrain using a noise function and a height map as one of the procedural terrain generation methods. The polygon mesh data structure to represent the generated terrain concisely and render it is also described. The Perlin noise function is used as the noise technique for terrain mesh, and the height data of the terrain is generated by combining the four noise waves. In addition, the terrain height information can be also obtained from actual image data taken from the satellite. The algorithm presented in this paper generates the geometry part of the polygon topography from the height data obtained, and generated a material for texture mapping with two textures, that is, a diffuse texture and a normal texture. The validity of the terrain method proposed in this paper is verified through application examples, and its possibility can be confirmed through performance verification.

• 한국멀티미디어학회논문지
• /
• 제18권9호
• /
• pp.1008-1018
• /
• 2015

In this paper, we propose a method that constructs a saliency map in which important regions are accurately specified and the colors of the regions are less influenced by the similar surrounding colors. Our method utilizes LBP(Local Binary Pattern) histogram information to compare and analyze texture information of surrounding regions in order to reduce the effect of color information. We extract the saliency of stereoscopic images by integrating a 2D saliency map with depth information of stereoscopic images. We then measure the distance between two different sizes of the LBP histograms that are generated from pixels. The distance we measure is texture difference between the surrounding regions. We then assign a saliency value according to the distance in LBP histogram. To evaluate our experimental results, we measure the F-measure compared to ground-truth by thresholding a saliency map at 0.8. The average F-Measure is 0.65 and our experimental results show improved performance in comparison with existing other saliency map extraction methods.

• 한국의류학회지
• /
• 제30권5호
• /
• pp.699-710
• /
• 2006

The objective of this study was to identify the texture-related components of woven fabrics and to develop a multidimensional perceptual structure map to represent the tactile textures. Eighty subjects in clothing and tektite industries were selected for multivariate data on each fabric of 30 using the questionnaire with 9 pointed semantic differential scales of 20 texture-related adjectives. Data were analyzed by factor analysis, hierarchical cluster analysis, and multidimensional scaling(MDS) using SPSS statistical package. The results showed that the five factors were selected and composed of density/warmth-coolness, stiffness, extensibility, drapeability, and surface/slipperiness. As a result of hierarchical cluster analysis, 30 fabrics were grouped by four clusters; each cluster was named with density/warmth-coolness, surface/slipperiness, stiffness, and extensibility, respectively. By MDS, three dimensions of tactile texture were obtained and a 3-dimensional perceptual structure map was suggested. The three dimensions were named as surface/slipperiness, extensibility, and stiffness. We proposed a positioning perceptual map of fabrics related to texture naming system(TNS). To classify the textural features of the woven fabrics, hierarchical cluster analysis containing all the data variations, even though it includes the errors, may be more desirable than texture-related multidimensional data analysis based on factor loading values in respect of the effective variables reduction without losing the critical variations.

• Journal of Electrical Engineering and Technology
• /
• 제10권5호
• /
• pp.2189-2196
• /
• 2015

Stereo matching algorithm is usually used to obtain a disparity map from a pair of images. However, the disparity map obtained by using stereo matching contains lots of noise and error regions. In this paper, we propose a virtual-view synthesis algorithm using disparity refinement in order to improve the quality of the synthesized image. First, the error region is detected by examining the consistency of the disparity maps. Then, motion information is acquired by applying optical flow to texture component of the image in order to improve the performance. Then, the occlusion region is found using optical flow on the texture component of the image in order to improve the performance of the optical flow. The refined disparity map is finally used for the synthesis of the virtual view image. The experimental results show that the proposed algorithm improves the quality of the generated virtual-view.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2003년도 Proceedings of ACRS 2003 ISRS
• /
• pp.744-746
• /
• 2003

A 3D modeling of urban area can be composed the terrain modeling that can express specific and shape of the terrain and the object modeling such as buildings, trees and facilities which are found in urban areas. Especially in a 3D modeling of building, it is very important to make a unit model by simplifying 3D structure and to take a texture mapping, which can help visualize surface information. In this study, the texture mapping technique, based on library for 3D urban modeling, was used for building modeling. This technique applies the texture map in the form of library which is constructed as building types, and then take mapping to the 3D building frame. For effectively apply, this technique, we classified buildings automatically using LiDAR data and made 3D frame using LiDAR and digital map. To express the realistic building texture, we made the texture library using real building photograph.

• 방송공학회논문지
• /
• 제28권3호
• /
• pp.314-328
• /
• 2023

본 논문에서는 동적 메쉬 부/복호화 시 스케일러빌리티 기능을 지원하기 위해 SHVC의 계층적 부호화 방식을 기반으로 텍스처 맵을 압축하는 방법을 제안한다. 제안하는 방법은 고해상도 텍스처 맵을 다운샘플링하여 다해상도의 텍스처 맵을 생성하고 이를 SHVC로 부호화함으로써 효과적으로 다해상도 텍스처 맵들의 중복성을 제거한다. 동적 메쉬 복호화기에서는 수신기 성능, 네트워크 환경 등에 따라 적합한 해상도의 텍스처 맵을 복호화하여 메쉬 데이터의 스케일러빌리티를 지원할 수 있도록 한다. 제안하는 방법의 성능을 검증하기 위해 V-DMC (Video-based Dynamic Mesh Coding) 참조 소프트웨어인 TMMv1.0에 제안하는 방법을 적용하고 본 논문에서 제안하는 스케일러블 부/복호화기와 TMMv1.0 기반의 시뮬캐스트 방식의 성능을 비교하였다. 제안하는 방법은 시뮬캐스트 방법 대비 AI, LD 환경에서 Luma BD-rate (Luma PSNR)가 각각 평균 -7.7%, -5.7%의 향상된 결과를 얻어 제안하는 방법을 통해 효과적으로 동적 메쉬 데이터의 텍스처 맵 스케일러빌리티 지원이 가능함을 확인하였다.

• 한국융합학회논문지
• /
• 제3권1호
• /
• pp.13-17
• /
• 2012

본 논문에서는 기하학적 단서인 소실선과 텍스처를 이용하여 깊이 지도를 생성하는 방법을 제안한다. 소실선은 영상 내 존재하는 평행한 직선들에 의해 생성되는 것으로 영상에서 Gabor Filter를 통해 특정 각도의 경계를 추출하고 이를 허프 변환을 통해 직선을 추출하여 소실선을 검출해낸다. 검출된 소실선에 따라 초기 깊이 지도를 생성하고 텍스처 단서인 슈퍼 픽셀을 이용한 상대적 깊이 지도를 결합하여 최종 깊이 지도를 생성한다. 소실선을 이용한 초기 깊이지도와 슈퍼 픽셀을 이용한 상대적 깊이 지도를 결합함으로써 보다 신뢰성 있는 깊이 지도가 생성되었다.