• Proceedings of the Korea Multimedia Society Conference
• /
• 2001.11a
• /
• pp.299-302
• /
• 2001

이미지 검색 시스템이란 이미지가 갖는 다양한 특징을 바탕으로 똑같거나 유사한 이미지를 검색하여 제공하는 시스템이다. 본 논문에서는 이미지의 컬러와 형태를 기반으로 한 브랜드 이미지 검색 시스템을 제시한다. 이미지를 영역별로 분할하여 영역별 컬러 분포 히스토그램을 추출하여 컬러 정보로 이용하고 경계면 추출, 무게 중심 추출, angular 샘플링 등의 전처리 과정과 무게 중심으로부터 경계면까지 거리의 합, 표준 편차, 장/단축 비율을 계산하여 형태정보로 이용한다. 이렇게 추출된 컬러와 형태 정보를 이용하여 유사성 측정을 통한 검색을 수행한다.

• Journal of Korea Multimedia Society
• /
• v.14 no.4
• /
• pp.529-537
• /
• 2011

In this paper, we propose a stable scale adaptive tracking method that uses centroids of the target colors. Most scale adaptive tracking methods have utilized histograms to determine target window sizes. However, in certain cases, histograms fail to provide good estimates of target sizes, for example, in the case of occlusion or the appearance of colors in the background that are similar to the target colors. This is due to the fact that histograms are related to the numbers of pixels that correspond to the target colors. Therefore, we propose the use of centroids that correspond to the target colors in the scale adaptation algorithm, since centroids are less sensitive to changes in the number of pixels that correspond to the target colors. Due to the spatial information inherent in centroids, a direct relationship can be established between centroids and the scale of target regions. Generally, after the zooming factors that correspond to all the target colors are calculated, the unreliable zooming factors are filtered out to produce a reliable zooming factor that determines the new scale of the target. Combined with the centroid based tracking algorithm, the proposed scale adaptation method results in a stable scale adaptive tracking algorithm. It tracks objects in a stable way, even when the background colors are similar to the colors of the object.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.3
• /
• pp.167-172
• /
• 2006

An image retrieval system retrieves and offers same of similar image based on various features of image. This paper present a brand image retrieval system based on color and shape of image. We use the image for a color information by dividing into the area and extracting the area color distribution histogram. We use for the shape information by preprocessing of the boundary extraction, the centroid extraction, angular sampling etc. and calculating of the sum of the distance from the centroid to the boundary, the standard deviation, and the rate of long axis to short axis. We accomplish the retrieval through a similarity measurement by using the color and shape information which is extracted in this way.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.10b
• /
• pp.867-870
• /
• 2000

영상에서 나타난 컬러 정보를 이용해 검색을 수행하고자 하는 연구 방법에서 일반적으로 적용되어 사용되어진 방법 중 하나로 컬러 히스토그램의 유사성을 이용한 매핑을 들 수 있다. 이 방법의 장점은 쉽게 영상에서 나타난 컬러 정보를 그룹화하여 매핑 할 수 있게 해주나, 전체 영상에서 나타난 컬러의 양 정보만을 제공할 뿐, 각각의 컬러에 대한 공간적 정보를 제공하지는 못한다. 즉, 동일한 컬러가 한 곳에 집중해서 나타난 영상과 분산되어 나타난 영상이 동일한 컬러 정보로 매핑되므로 원치 않는 검색 결과를 얻게 된다. 이러한 문제점을 해결하기 위해 기존의 방법은 컬러 히스토그램에서 얻어진 컬러 정보와 별개의 공간 정보 추출 알고리즘을 통해 얻어진 정보를 결합(fusion) 하여 문제를 해결하였다. 본 연구에서는 이러한 단점을 해결하기 위해 각각의 히스토그램의 빈(bin)들에 컬러에 대한 양 정보만이 아니라 컬러의 중심점과 분산 값 정보를 구하여 이를 영상 검색을 위한 매핑 정보로 사용한다.

• The KIPS Transactions:PartB
• /
• v.9B no.4
• /
• pp.501-508
• /
• 2002

In this paper, we propose a content-based image retrieval using the color ratio and moment of object region. We acquire an optimal spatial information by the region splitting that utilizes horizontal-vertical projection and dominant color. It is based on hypothesis that an object locates in the center of image. We use color ratio and moment as feature informations. Those are extracted from the splitted regions and have the invariant property for various transformation, and besides, similarity measure utilizes a modified histogram intersection to acquire correlation information between bins in a color histogram. In experimental results, the proposed method shows more flexible and efficient performance than existing methods based on region splitting.

• Proceedings of the Korea Society of Design Studies Conference
• /
• 2001.05a
• /
• pp.110-111
• /
• 2001

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2010.05a
• /
• pp.160-161
• /
• 2010

In this paper, trademark image retrieval system is proposed by using color information and shape information. We use the image for a color information by dividing into the area and extracting the area color distribution histogram. We use for the shape information by preprocessing of the boundary extraction, the centroid extraction, angular sampling etc. and calculating of the sum of the distance from the centroid to the boundary, the standard deviation, and the rate of long axis to short axis. In particular, centroid by using the angular sampling can extract feature and reduce the processing time. Users can perform searchs using the color and shape information, and also the two methods by mixing can be used by weighting.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.9
• /
• pp.76-90
• /
• 1999

In this paper, a color image retrieval algorithm is proposed based on color histogram and color texture. The representative color vectors of a color image are made from k-means clustering of its color histogram, and color texture is generated by centering around the color of pixels with its color vector. Thus the color texture means texture properties emphasized by its color histogram, and it is analyzed by Gaussian Markov Random Field (GMRF) model. The proposed algorithm can work efficiently because it does not require any low level image processing such as segmentation or edge detection, so it outperforms the traditional algorithms which use color histogram only or texture properties come from image intensity.

• Journal of Broadcast Engineering
• /
• v.16 no.1
• /
• pp.178-181
• /
• 2011

With the development of mobile devices and intelligent surveillance system loaded with pan/tilt cameras, object tracking with non-stationary cameras has become a topic with increasing importancy. Since it is difficult to model a background image in a non-stationary camera environment, colors and texture are the most important features in the tracking algorithm. However, colors in the background similar to those in the target arise instability in the tracking. Recently, we proposed a robust color based tracking algorithm that uses an area weighted centroid shift. In this letter, we update the model such that it becomes more stable against background colors. The proposed algorithm also incorporates time filtering by adding an additional energy term to the energy functional.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2003.05b
• /
• pp.488-491
• /
• 2003

본 논문에서는 실시간 CCD 카메라 입력 영상으로부터 다중 컬러 정보를 이용하여 얼굴 영역을 검출 및 추적하고 기울어진 얼굴을 보정하는 알고리즘을 제안하였다. 제안한 알고리즘은 먼저 획득된 RGB 영상에서 YCbCr컬러 모델과 YIQ컬러 모델로 변환한 후 Cr성분과 I성분을 추출하여 얼굴 피부색을 검출, 얼굴 영역 추출에 사용하였다. 또한 추출된 얼굴 후보 영역에서 수평, 수직 투영(Projection)정보로부터 최종 얼굴 영역으로 검출한 다음 검출된 얼굴 중심 좌표와 이전에 검출된 얼굴 중심 좌표 값을 유클리드언 거리로 얼굴을 추적하였으며 검출된 얼굴로부터 레이블링(Labeling)기법으로 눈 특징자를 검출, 눈의 기울기 각도를 보정함으로써 얼굴 기울기를 보정하였다. 제안한 얼굴 추적 및 기울기 보정 알고리즘을 사용하여 실험한 결과 다중 색상 정보를 사용함으로써 주위환경 변화에 강인하게 실시간 얼굴 영역 김출 및 추적이 가능하였고, 기울어진 얼굴 영상을 자동 보정함으로써 인식에 용이하였다.