• 전기전자학회논문지
• /
• 제6권1호
• /
• pp.54-64
• /
• 2002

본 논문에서는 화상회의 영상 데이터에서 화자를 분할하고, 그 움직임을 추적하는 알고리즘을 제안하였다. 실시간 처리가 가능하도록 화자 분할과 움직임 추적 순으로 처리되는 알고리즘으로 단순화하였다. 분할 한계에서는 차분 방법에 의해 구한 움직임 정보와 영상의 밝기 정보를 사용하여 화자를 분할하였다. 분할된 화자로부터 기준 마스크 영상을 생성하였다. 움직임 추적 단계에서는 움직임 추적에 불필요한 블록들은 제외함으로써 빠르게 움직임을 추적할 수 있는 블록정합 알고리즘을 사용하여 추적하였다. 시뮬레이션에서 여러 시험 영상에 제안한 알고리즘을 적용하여 움직이는 화자를 분할하고, 그 움직임를 추적하는 올바른 결과를 얻을 수 있었다.

• 통합자연과학논문집
• /
• 제11권3호
• /
• pp.131-138
• /
• 2018

In this paper, we have implemented speeches that utilized the emotion information of the user's speech and image matching and recommendation system. To classify the user's emotional information of speech, the emotional information of speech about the user's speech is extracted and classified using the PLP algorithm. After classification, an emotional DB of speech is constructed. Moreover, emotional color and emotional vocabulary through factor analysis are matched to one space in order to classify emotional information of image. And a standardized image recommendation system based on the matching of each keyword with the BM-GA algorithm for the data of the emotional information of speech and emotional information of image according to the more appropriate emotional information of speech of the user. As a result of the performance evaluation, recognition rate of standardized vocabulary in four stages according to speech was 80.48% on average and system user satisfaction was 82.4%. Therefore, it is expected that the classification of images according to the user's speech information will be helpful for the study of emotional exchange between the user and the computer.

• 융합신호처리학회논문지
• /
• 제7권4호
• /
• pp.155-160
• /
• 2006

본 논문에서는 웨이브릿 변환 영역의 4개의 서브밴드와 임계값 처리를 이용하는 계층적 블록 매칭 알고리즘을 제안하였다. 제안한 알고리즘은 다분해능의 첫 번째 레벨에서 웨이브릿 변환 영역의 4개 서브밴드를 이용함으로써 복원된 영상의 PSNR 성능을 향상시켰고, 움직임 벡터에 대한 임계값 처리를 하여 계산량을 줄였다. Carphone 영상과 Mother and Daughter 영상에 대한 실험에서 기존의 계층적 움직임 추정 알고리즘과 비교하여 임계값을 0으로 하였을 경우에 계산량은 최대 16%까지 증가하였으나 복원된 영상의 PSNR 성능은 최대 0.16dB 정도 향상된 결과를 보였고, 임계값을 증가시킴에 따라서 계산량은 최대 8%정도 줄고 복원된 영상의 PSNR은 비슷한 성능을 보였다.

• 융합정보논문지
• /
• 제7권2호
• /
• pp.77-82
• /
• 2017

본 논문은 차량영상을 입력영상으로 받아 차량의 종류를 인식하는 보안 알고리즘을 연구한다. 차량 인식 보안 알고리즘은 영상입력, 배경제거, 에지영역 추출, 전처리(이진화), 차량인식 등의 5가지 핵심부분으로 구성된다. 그러므로 차량 종류 인식 보안 알고리즘의 최종 인식율은 각 단계의 역할 및 기능에 직간접적인 영향을 받는다. 영상을 그레이 스케일 이미지로 입력시킨 후 배경을 제거하고, 에지영역만 추출한 후 이진화를 거친다. 외곽선을 또렷하게 해주기 위한 전처리 과정을 거친 후 차량의 높이와 넓이의 비율을 통해 차량의 종류를 대형차, 승용차, 오토바이의 3가지 범주로 나누게 했다.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 1996년도 학술대회
• /
• pp.63-66
• /
• 1996

In this paper, we propose a new content-based color image retrieval algorithm. The algorithm makes use of two features; colors as global features and block classification results as local features. More specifically, we obtain R, G, B color histograms and classify nonoverlapping small image blocks into texture, monotone, and various edges, then using these histograms and classification results were make a similarity measure. Experimental results show that retrieval rate of the proposed algorithm is higher than the previous method.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1998년도 추계종합학술대회 논문집
• /
• pp.895-898
• /
• 1998

In this paper, we propose an efficient marker extraction algorithm for initial image segmentation in a bottom-up segmentation scheme. The proposed algorithm generates dense markers in visually complex areas and coarse markers in visually uniform areas. which conforms to the human perceptual system. Experimental results show that the proposed method achieves better subjective quality for fine initial image segmentation.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1998년도 The Third Asian Fuzzy Systems Symposium
• /
• pp.479-484
• /
• 1998

As technology in a computer hardware and software advances, efficient information retrieval from multimedia database gets highly demanded. Recently, it has been actively exploited to retrieve information based on the stored contents. However, most of the methods emphasize on the points which are far from human intuition or emotion. In order to overcome this shortcoming , this paper attempts to apply interactive genetic algorithm to content-based image retrieval. A preliminary result with subjective test shows the usefulness of this approach.

• 한국멀티미디어학회논문지
• /
• 제8권12호
• /
• pp.1565-1571
• /
• 2005

Images reconstructed with Maximum-Likelihood Expectation-Maximization (MLEM) algorithm have been observed to have checkerboard effects and have noise artifacts near edges as iterations proceed. To compensate this ill-posed nature, numerous penalized maximum-likelihood methods have been proposed. We suggest a simple algorithm of applying edge detecting process to the MLEM and Bayesian Expectation-Maximization (BEM) to reduce the noise artifacts near edges and remove checkerboard effects. We have shown by simulation that this algorithm removes checkerboard effects and improves the clarity of the reconstructed image and has good properties based on root mean square error (RMS).

• 한국진공학회지
• /
• 제8권3B호
• /
• pp.315-321
• /
• 1999

In this paper, the SEM image processing system based on PC is designed, and a new noise reduction filtering algorithm is proposed. The SEM image obtained in semiconductor processing line is sensitive to noise, the weighted-D filter can remove uniform and Gaussian noise effectively, but can not remove impulse noise properly, A new improved filtering algorithm is proposed to reduce mixed-noise. The performance of the proposed filter is quantitatively evaluated by use of the normalized mean square errors (NMSE). The experimental results show that the performance of the proposed filter is obtained between 0.96 and 2.5 times better than that of weighted-D filter in NMSE evaluation.

• Journal of information and communication convergence engineering
• /
• 제9권2호
• /
• pp.197-200
• /
• 2011

In this paper, we propose a multi-view stereoscopic image synthesis algorithm for 3DTV system using depth information with an RGB texture from a depth camera. The proposed algorithm synthesizes multi-view images which a virtual convergence camera model could generate. Experimental results showed that the performance of the proposed algorithm is better than those of conventional methods.