• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.11a
• /
• pp.96-99
• /
• 2008

본 논문은 의료 영상을 효과적으로 분류하고 검색 하기 위한 새로운 알고리즘을 제안한다. 의료 영상 중 X-Ray 영상은 어두운 배경에 반해 밝은 전경을 갖고 있기 때문에, 전경의 두드러진 부분에서만 시각 기술자로 추출한다. 우선, 색 구조 기술자(H-CSD)에서 해리스 코너 검출기로 검출한 관심 포인트들에서 색상 특징을 추출하고, 경계선 히스토그램 기술자에서 영상의 전역 및 지역적 질감 특징을 추출한다. 추출된 특징 벡터는 멀티클래스 SVM 에 적용되어 각 영상을 위한 멤버십 스코어를 얻는다. 이후, H-CSD와 EHD 에 대한 SVM 의 멤버십 스코어를 앙상블 결합하여 하나의 특징 벡터로 생성하고, K-nearest Neighborhood 방법을 이용하여 상위-K 개의 영상을 검색을 하도록 하였다. imageCLEFmed2007 을 이용한 실험 결과에서 다른 전역적 속성 또는 분류 기반 검색 방법에 비교하여 보다 개선된 검색 성능을 나타냄을 확인하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.676-678
• /
• 2005

디지털 비디오에 있어서 멀티미디어 데이터베이스 및 검색 시스템 구축을 하기 위해서는 비디오를 여러개의 장면으로 분할하는 기술이 요구된다. 본 논문에서는 다양한 종류의 장면 전환을 검출하기 위해 기존의 규칙기반이 아닌 신경망 이론을 접목하여 자율학습과 실수값 입력이 가능한 ART2를 이용하였다. 매프레임마다 발생할 수 있는 변동의 폭을 줄이기 위해 MPEG 동영상의 DC에 해당하는 값만을 이용하고, 프레임마다 색상의 분산값을 이용하여 Plateaus 구간을 검출한 다음 Plateaus 구간에 해당하는 프레임들에 대해서만 프레임차이, 히스토그램차이, 상관계수 등의 특징치를 추출하여 ART2에 특징벡터를 입력하여 장면 전환을 검출하였다.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.3
• /
• pp.52-59
• /
• 2009

In general, methods based on histogram or a correction of gamma curve are usually utilized to enhance the contrast of captured image in the dark scene. These methods are efficient to enhance the contrast globally, however, they locally induced the low quality of image. Recently, to resolve the problem, the multi-scaled refiner algorithm improving the contrast with locally averaged lightness is proposed. However, estimating the locally averaged lightness, if there is the object with a high saturated color, the color distortion might be induced by the color of object. Thus, in this paper, the dominant chromaticity of image is estimated to correct the locally averaged lightness in multi-scaled retinex algorithm. Because the average chromaticity of image includes the chromaticity of illumination, the dominant chromaticity is estimated with dividing the average chromaticity of image by the estimated chromaticity of illumination from highlight region. In addition, to improve the lower chroma by multi-scaled retinex algorithm generally, the chroma was compensated preserving the hue in the CIELAB color space.

• The Journal of the Korea Contents Association
• /
• v.5 no.4
• /
• pp.150-159
• /
• 2005

With the increase of interest in ubiquitous computing, image code is attracting attention in various areas. Image code is important in ubiquitous computing in that it can complement or replace RFID (radio frequency identification) in quite a few areas as well as it is more economical. However, because of the difficulty in reading precise colors due to the severe distortion of colors, its application is quite restricted by far. In this paper, we present an efficient method of image code recognition including automatically locating the image code using the hue and saturation values. In our experiments, we use an image code whose design seems most practical among currently commercialized ones. This image code uses six safe colors, i.e., R, G, B, C, M, and Y. We tested for 72 true-color field images with the size of $2464{\times}1632$ pixels. With the color calibration based on the histogram, the localization accuracy was about 96%, and the accuracy of color classification for localized codes was about 91.28%. It took approximately 5 seconds to locate and recognize the image code on a PC with 2 GHz P4 CPU.

• Science of Emotion and Sensibility
• /
• v.7 no.4
• /
• pp.57-65
• /
• 2004

In this paper, we present a hybrid approach which incorporates color, texture and shape in content-based image retrieval. Colors in each image are clustered into a small number of representative colors. The feature descriptor consists of the representative colors and their percentages in the image. A similarity measure similar to the cumulative color histogram distance measure is defined for this descriptor. The co-occurrence matrix as a statistical method is used for texture analysis. An optimal set of five statistical functions are extracted from the co-occurrence matrix of each image, in order to render the feature vector for eachimage maximally informative. The edge information captured within edge histograms is extracted after a pre-processing phase that performs color transformation, quantization, and filtering. The features where thus extracted and stored within feature vectors and were later compared with an intersection-based method. The content-based retrieval system is tested to be effective in terms of retrieval and scalability through experimental results and precision-recall analysis.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.44 no.1
• /
• pp.40-49
• /
• 2007

In a visual driver-assistance system, separating moving objects from fixed objects are an important problem to maintain multiple hypothesis for the state. Color and edge-based tracker can often be 'distracted' causing them to track the wrong object. Many researchers have dealt with this problem by using multiple features, as it is unlikely that all will be distracted at the same time. In this paper, we improve the accuracy and robustness of real-time tracking by combining a color histogram feature with a brightness of Optical Flow-based feature under a Sequential Monte Carlo framework. And it is also excepted from Tracking as time goes on, reducing density by Adaptive Particles Number in case of the fixed object. This new framework makes two main contributions. The one is about the prediction framework which separating moving objects from fixed objects and the other is about measurement framework to get a information from the visual data under a partial occlusion.

• The KIPS Transactions:PartB
• /
• v.11B no.2
• /
• pp.149-158
• /
• 2004

This study proposes a method for segmentation and recognition of traffic signs using shape information and edge image in real image. It first segments traffic sign candidate regions by connected component algorithm from binary images, obtained by utilizing the RGB color ratio of each pixel in the image, and then extracts actual traffic signs based on their symmetries on X- and Y-axes. Histogram equalization is performed for unsegmented candidate regions caused by low contrast in the image. In the recognition stage, it utilizes shape information including projection profiles on X- and Y-axes, moment, and the number of crossings and distance which concentric circular patterns and 8-directional rays from region center intersects with edges of traffic signs. It finally performs recognition by measuring similarity with the templates in the database. It will be shown from several experimental results that the system is robust to environmental factors, such as light and weather condition.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.12
• /
• pp.168-175
• /
• 2013

Color consistency in stereoscopic contents is important for 3D display systems. Even with a stereo camera of the same model and with the same hardware settings, complex color discrepancies occur when acquiring high quality stereo images. In this paper, we propose an integrated color matching method that use cumulative histogram in global matching and estimated 3D-distance for the stage of local matching. The distance between the current pixel and the target local region is computed using depth information and the spatial distance in the 2D image plane. The 3D-distance is then used to determine the similarity between the current pixel and the target local region. The overall algorithm is described as follow; First, the cumulative histogram matching is introduced for reducing global color discrepancies. Then, the proposed local color matching is established for reducing local discrepancies. Finally, a weight-based combination of global and local matching is computed. Experimental results show the proposed algorithm has improved global and local error correction performance for stereoscopic contents with respect to other approaches.

• Journal of KIISE:Computing Practices and Letters
• /
• v.10 no.6
• /
• pp.514-528
• /
• 2004

An emotion-based video scene retrieval algorithm is proposed in this paper. First, abrupt/gradual shot boundaries are detected in the video clip representing a specific story Then, five video features such as 'average color histogram' 'average brightness', 'average edge histogram', 'average shot duration', and 'gradual change rate' are extracted from each of the videos and mapping between these features and the emotional space that user has in mind is achieved by an interactive genetic algorithm. Once the proposed algorithm has selected videos that contain the corresponding emotion from initial population of videos, feature vectors from the selected videos are regarded as chromosomes and a genetic crossover is applied over them. Next, new chromosomes after crossover and feature vectors in the database videos are compared based on the similarity function to obtain the most similar videos as solutions of the next generation. By iterating above procedures, new population of videos that user has in mind are retrieved. In order to show the validity of the proposed method, six example categories such as 'action', 'excitement', 'suspense', 'quietness', 'relaxation', 'happiness' are used as emotions for experiments. Over 300 commercial videos, retrieval results show 70% effectiveness in average.

• 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.