• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.223-228
• /
• 2009

We propose an on-line machine learning approach for object recognition, where new images are continuously added and the recognition decision is made without delay. Random forest (RF) classifier has been extensively used as a generative model for classification and regression applications. We extend this technique for the task of building incremental component-based detector. First we employ object descriptor model based on bag of covariance matrices, to represent an object region then run our on-line RF learner to select object descriptors and to learn an object classifier. Experiments of the object recognition are provided to verify the effectiveness of the proposed approach. Results demonstrate that the propose model yields in object recognition performance comparable to the benchmark standard RF, AdaBoost, and SVM classifiers.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.396-399
• /
• 2009

In this paper, we present a method to define a color similarity between color images using Octree-based quantization and similar color integration. The proposed method defines major colors from each image using Octree-based quantization. Two color palettes to consist of major colors are compared based on Euclidean distance and similar color bins between palettes are matched. Multiple matched color bins are integrated and major colors are adjusted. Color histogram based on the color palette is constructed for each image and the difference between two histograms is computed by the weighted Euclidean distance between the matched color bins in consideration of the frequency of each bin. As an experiment to validate the usefulness, we discriminated the same clothing from CCD camera images based on the proposed color similarity analysis. We retrieved the same clothing images with the success rate of 88 % using only color analysis without texture analysis.

• International Journal of Contents
• /
• 제5권4호
• /
• pp.51-54
• /
• 2009

In huge video databases, the effective video content indexing method is required. While manual indexing is the most effective approach to this goal, it is slow and expensive. Thus automatic indexing is desirable and recently various indexing tools for video databases have been developed. For efficient video content indexing, the similarity measure is an important factor. This paper presents new similarity measures between frames and proposes a new algorithm to index video content using Kullback-Leibler distance defined between two histograms. Experimental results show that the proposed algorithm using Kullback-Leibler distance gives remarkable high accuracy ratios compared with several conventional algorithms to index video content.

• 방송공학회논문지
• /
• 제7권4호
• /
• pp.291-299
• /
• 2002

본 논문은 밝기 변화가 심한 비디오 시퀀스에 대해 효율적인 움직임 보상 알고리즘을 제안한다 제안한 알고리즘에서는 화면간의 밝기 변화 변수들을 추정하고 지역적인 움직임 보상을 수행한다. 밝기 변화가 심한 화면을 검출하기 위해 화면간의 히스토그램의 크로스 엔트로피를 계산하여 밝기 변화가 심한 화면을 그렇지 않은 화면과 나누어 불필요한 계산량을 줄였다. 밝기 변화가 심한 비디오 시퀀스에 대한 실험결과 제안한 알고리즘은 기존의 알고리즘에 비해 적은 계산량으로 높은 PSNR (peak signal to noise ratio) 성능을 나타내었다.

• 제어로봇시스템학회논문지
• /
• 제7권4호
• /
• pp.304-308
• /
• 2001

The increase of video data makes the demand of efficient retrieval, storing, and browsing technologies necessary. In this paper, a video segmentation method (scene change detection method, or shot boundary detection method) for the development of such systems is proposed. For abrupt cut detection, inter-frame similarities are computed using luminance and edge histograms and a cut is declared when the similarities are under th predetermined threshold values. A gradual scene change detection is based on the similarities between the current frame and the previous shot boundary frame. A correlation method is used to obtain universal threshold values, which are applied to various video data. Experimental results show that propose method provides 90% precision and 98% recall rates for abrupt cut, and 59% precision and 79% recall rates for gradual change.

• Communications for Statistical Applications and Methods
• /
• 제2권2호
• /
• pp.176-183
• /
• 1995

Mean opinion score (MOS) method has been used in many areas to quantify opinions of respondents not only in survey research but in evaluating the parameters of population that are not measurable of are technically hard to be measured. Histogram is an important graphical technique because of the role it plays in describing categorical data as well as quantitative. In MOS method, subjective opinions of respondents are quantified by opinion scores and the arithmetic means of opinion scores have been used to describe the interesting population. Since opinion scores are polytomous, the values of arithmetic means have little meanings. In this paper, cumulative percentage curves as a function of the means of opinion scores are derived by combining means of opinion scores and histograms. It is proposed for better interpretation to opinion scores in MOS method, one of subjective evaluation methods.

• 대한기계학회논문집A
• /
• 제32권11호
• /
• pp.963-969
• /
• 2008

The wear particles are relative to the failure and the life of machine elements directly. But it is not laid down to calculate shape parameters of wear particle. To analyze a variation of distributed characteristics of wear particles on moving conditions, its shape parameters such as diameter and roundness were calculated the quantitative values by digital image processing, and had to be defined the effective method of using those data. Up to the present, the shape parameters have been used simply into the average values. But these values are not effective to analyze a variation of distributed characteristics of occurred wear particles on moving conditions. In this study, the relative histograms of shape parameters of wear particles were used for the purpose of analyzing the distribution of wear particles in various conditions. The results showed that the relative histogram of shape parameters can be effectively represented to study a wear mechanism.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권11호
• /
• pp.4534-4555
• /
• 2015

We propose a robust object cataloging method using multiple locally distinct heterogeneous features for aiding image retrieval. Due to challenges such as variations in object size, orientation, illumination etc. object recognition is extraordinarily challenging problem. In these circumstances, we adapt local interest point detection method which locates prototypical local components in object imageries. In each local component, we exploit heterogeneous features such as gradient-weighted orientation histogram, sum of wavelet responses, histograms using different color spaces etc. and combine these features together to describe each component divergently. A global signature is formed by adapting the concept of bag of feature model which counts frequencies of its local components with respect to words in a dictionary. The proposed method demonstrates its excellence in classifying objects in various complex backgrounds. Our proposed local feature shows classification accuracy of 98% while SURF,SIFT, BRISK and FREAK get 81%, 88%, 84% and 87% respectively.

• 한국측량학회지
• /
• 제9권2호
• /
• pp.115-118
• /
• 1991

토지 정보 시스템은 토지의 물리적, 법적 특성을 데이타 베이스화하는 것으로서 토지 기록 시스템이라 부르기도 한다. 수치원격탐사 자료는 스펙트럴 자료의 기본적인 통계를 계산하는데 유용하다. 각 밴드의 최대, 최소값, 평균, 표준편차, 분산-공분산 행렬, 그리고 밝기값의 빈도수를 막대도표로 그릴 수 있다. 본 논문에서는 항공사진을 스캔닝 하였으며, 다음 논문에 인공위성 영상을 스캔닝 할 것이다.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2011년도 추계학술대회
• /
• pp.169-171
• /
• 2011

최근 통신 환경의 진화와 고사양 모바일 단말기의 대중화로 인하여 모바일 통신 환경에서 동영상 서비스를 이용하는 사용자가 급증하고 있다. 사용자들은 제한적인 데이터 사용량을 이용하여 필요한 영상을 효율적으로 검색 및 탐색하기 위하여 미리보기 서비스 또는 Thumbnail을 이용한다. 이 중 Thumbnail은 동영상의 복호화 과정에서 제공되는 서비스이며, 이를 효과적으로 추출하는데 있어서 낮은 복잡도의 기술들이 사용되어야 한다. 본 논문에서는 영상의 히스토그램의 특징을 이용한 Thumbnail 추출 기법을 제안한다. 제안하는 추출 기법에서는 각 영상의 히스토그램을 비교 및 분석하여 하나의 video을 다수의 구간으로 분할하고, 각 구간의 대표 Thumbnail을 추출하는 방법을 사용한다. 다양한 영상을 이용한 실험 결과는 제안하는 알고리즘이 효과적으로 Thumbnail을 추출함을 보여준다.