• KSCI Review
• /
• v.14 no.1
• /
• pp.255-264
• /
• 2006

Video viewers can not understand enough entire video contents because most video is long length data of large capacity. This paper propose efficient scene change detection and video abstracting using new shot clustering to solve this problem. Scene change detection is extracted by method that was merged color histogram with ${\chi}^2$ histogram. Clustering is performed by similarity measure using difference of local histogram and new shot merge algorithm. Furthermore, experimental result is represented by using Real TV broadcast program.

• Journal of the Korea Society of Computer and Information
• /
• v.12 no.2 s.46
• /
• pp.103-109
• /
• 2007

In this paper, we detect shot boundaries using $X^2$-histogram comparison method which have enough spatial information that is more robust to the camera or object motion and produce more precise results. Also, we present normalization method to change Log-Formula and constant that is used for contrast enhancement of image in image processing and apply in difference value. And, present shot boundary detection algorithm to detect shot boundary based on general shot and abrupt shot's characteristic.

• The KIPS Transactions:PartB
• /
• v.17B no.4
• /
• pp.275-286
• /
• 2010

Reversible watermarking which can preserve the original quality of the digital contents and protect the copyright has been studied actively. Especially, in medical, military, and art fields, the need for reversible watermarking is increasing. This paper proposes a high-capacity reversible watermarking through predicted error expansion and error estimation compensation. Watermark is embedded by expanding the difference histogram between the original value and the predicted value. Differently from previous methods calculating the difference between adjacent pixels, the presented method calculates the difference between the original value and the predicted value, and that increases the number of the histogram value, where the watermark is embedded. As a result, the high capacity is achieved. The inserted watermark is extracted by restoring the histogram between the original value and the predicted value. To prove the performance, the presented algorithm is compared with other previous methods on various test images. The result supports that the presented algorithm has a perfect reversibility, a high image quality, and a high capacity.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2002.11b
• /
• pp.141-144
• /
• 2002

영상에서의 칼라분포 정보는 영상간의 유사성을 표현하는데 매우 유용하여 내용기반 영상검색분야에서 기본적으로 사용하고 있다. 이때, 영상 데이터베이스에서의 각 영상에 대하여 동일한 방식으로 (비)균일하게 양자화하여 표현한 칼라 히스토그램이 주로 사용되고 있다. 그러나, 전체영상에 대하여 동일한 개수의 고정된 양자화를 통해 칼라분포 정보를 표현하는데, 여러 가지 문제점과 성능 차이가 있어 다양한 해결 방안이 연구되고 있다. 본 논문에서는, 적응적 양자화 방법으로 각 영상의 칼라분포 정보를 표현하되, 상이한 양자화 칼라간의 유사도를 정의하여 칼라히스토그램 인터섹션 방법과 유사하게 영상간의 칼라분포 유사도를 계산하는 방법을 제안한다. 양자화 칼라간의 유사도는 거리에 반비례하면서 두 양자화 칼라의 작은 빈도값에 비례하도록 정의하였다. 영상간의 칼라분포 유사도는 칼라 히스토그램 인터섹션 방법을 생산자-소비자 모델로 해석하여 구하는 방법을 제안한다. 제안한 방법에 의해 기존의 칼라 히스토그램 인터섹션 방법보다 향상된 결과를 얻을 수 있음을 실험을 통해 확인하였다.

• The KIPS Transactions:PartB
• /
• v.18B no.6
• /
• pp.355-364
• /
• 2011

In most applications requiring high-confidential images, reversible watermarking is an effective way to ensure the integrity of images. Many watermarking researches which have been adapted to authenticate contents cannot recover the original image after authentication. However, reversible watermarking inserts the watermark signal into digital contents in such a way that the original contents can be restored without any quality loss while preserving visual quality. To detect malicious tampering, this paper presents a new block-based image authentication algorithm using differential histogram-based reversible watermarking. To generate an authentication code, the DCT-based authentication feature from each image block is extracted and combined with user-specific code. Then, the authentication code is embedded into image itself with reversible watermarking. The image can be authenticated by comparing the extracted code and the newly generated code and restored into the original image. Through experiments using multiple images, we prove that the presented algorithm has achieved over 97% authentication rate with high visual quality and complete reversibility.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.6
• /
• pp.147-154
• /
• 2008

First, this paper compute difference value between frames using the composed method of $X^2$ histogram and color histogram and the normalization. Next, cluster representative frame was decided by using the clustering for distance and the k-mean grouping. Finally, representative frame of group was decided by using the likelihood ratio. Proposed method can be known by experiment as outstanding of detection rather than other methods, due to computing of difference value, clustering and grouping, and detecting of representative frame.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.10
• /
• pp.2427-2432
• /
• 2014

Color-$X^2$ is used as a method for scene change detection. It extracts a violent scene in an elevator and then could be used for real-time surveillance of criminal acts. The scene could be also used to secure after-discovered evidences and to prove analysis processes. Video Forensic is defined as a research on various methods to efficiently analyze evidences upon crime-related visual images in the field of digital forensic. The method to use differences of color-histogram detects the difference values of histogram for RGB color from two frames respectively. Our paper uses Color-$X^2$ histogram that is composed of merits of color histogram and ones of $X^2$ histogram, in order to efficiently extract violent scenes in elevator. Also, we use a threshold so as to find out key frame, by use of existing Color-$X^2$ histogram. To increase the probability that discerns whether a real violent scene or not, we take advantage of statistical judgments with 20 sample visual images.

• Journal of the Korea Society of Computer and Information
• /
• v.7 no.4
• /
• pp.75-82
• /
• 2002

Video viewers can not understand enough entire video contents because most video is long length data of large capacity. This paper Propose efficient scene change detection and video abstracting using new shot clustering to solve this problem. Scene change detection is extracted by method that was merged color histogram with χ2 histogram. Clustering is performed by similarity measure using difference of local histogram and new shot merge algorithm. Furthermore, experimental result is represented by using Real TV broadcast program.

• KIISE Transactions on Computing Practices
• /
• v.22 no.10
• /
• pp.532-536
• /
• 2016

Frame Rate Up Conversion (FRUC) generate interpolated frames between existing frames using motion estimation and motion compensation interpolation considering temporal redundancy. Falsely-estimated FRUC, however, can generate poor quality frames because FRUC typically uses blending-based interpolation method. As skipping an interpolating process between frames generate mis-estimated motion vectors, could improve Quality of Services of FRUC. In this Paper we analyze luminance histogram and motion vector consistency in frames generating poor quality interpolated frames. We conclude these features could help to be a clue in classifying the frames, which often result in the poor quality of FRUC results through the analysis and experiment.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.11a
• /
• pp.527-530
• /
• 2012

비디오 데이터를 효율적으로 검색, 정렬, 탐색, 분류하기 위해서는 프레임 간의 샷 전환 탐지가 선행되어야 한다. 본 논문에서는 디지털 비디오 데이터의 샷 전환 탐지를 위해 비디오 스트림을 구성하고 있는 각 프레임들 간의 화소 밝기 차이와 히스토그램의 변화를 이용하였다. 플래쉬 등과 같은 인위적이고 급격한 화소 밝기변화에 의한 오류를 최소화하기 위해 샷 전환 탐지 이전에 각 프레임 간의 밝기 보상을 적용하였다. 밝기 보정 된 프레임으로부터 프레임의 서브 블록 간의 지역적 화소 밝기 정보, 그리고 프레임의 화소 밝기 값 히스토그램을 비교하여 샷 전환을 탐지한다. 실험에서 제안된 알고리즘은 국가기록원 소장 비디오에 적용하여 효과가 있음을 보였다.