• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제18권9호
• /
• pp.2605-2625
• /
• 2024

Siamese-based segmentation and tracking algorithms improve accuracy and stability for video object segmentation and tracking tasks simultaneously. Although effective, variability in target appearance and background clutter can still affect segmentation accuracy and further influence the performance of tracking. In this paper, we present a memory propagation-based target-aware and mask-attention decision network for robust object segmentation and tracking. Firstly, a mask propagation-based attention module (MPAM) is constructed to explore the inherent correlation among image frames, which can mine mask information of the historical frames. By retrieving a memory bank (MB) that stores features and binary masks of historical frames, target attention maps are generated to highlight the target region on backbone features, thus suppressing the adverse effects of background clutter. Secondly, an attention refinement pathway (ARP) is designed to further refine the segmentation profile in the process of mask generation. A lightweight attention mechanism is introduced to calculate the weight of low-level features, paying more attention to low-level features sensitive to edge detail so as to obtain segmentation results. Finally, a mask fusion mechanism (MFM) is proposed to enhance the accuracy of the mask. By utilizing a mask quality assessment decision network, the corresponding quality scores of the "initial mask" and the "previous mask" can be obtained adaptively, thus achieving the assignment of weights and the fusion of masks. Therefore, the final mask enjoys higher accuracy and stability. Experimental results on multiple benchmarks demonstrate that our algorithm performs outstanding performance in a variety of challenging tracking tasks.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2008년도 하계종합학술대회
• /
• pp.1017-1018
• /
• 2008

In this paper, we propose a method that makes use of neighborhood relationship in 2D spectrogram of separated sources toward the generalization of the binary mask in Degenerate Unmixing Estimation Technique (DUET). A new generalized mask can be consist of five to ten mask. According to the new mask, the original power of the spectrogram in each frequency-time point is assigned. The result showed a smooth and tender wave-form, indicating a high speech separation performance compared to the original method.

• 한국광학회:학술대회논문집
• /
• 한국광학회 2000년도 하계학술발표회
• /
• pp.126-127
• /
• 2000

In this paper, we proposed an optical encryption method based in the concept of visual cryptography and interferometry. In our method a secret binary image was divided into two sub-images and they were encrypted by 'XOR' operation with a random key mask. Finally each encrypted image was changed into phase mask. By interference of these two phase masks the original image was obtained. Compared with general visual encryption method, this optical method had good signal-to-noise ratio due to no need to generate sub-pixels like visual encryption.

• 한국음향학회지
• /
• 제29권7호
• /
• pp.468-474
• /
• 2010

음성인식의 실용화에 가장 저해되는 요소는 배경잡음과 채널잡음에 의한 왜곡이다. 일반적으로 배경잡음은 음성인식 시스템의 성능을 저하시키고 이로 인해 사용 장소의 제약을 받게 한다. DSR (Distributed Speech Recognition) 기반의 음성인식 역시 이와 같은 문제로 성능 향상에 어려움을 겪고 있다. 이러한 문제를 해결하기 위해 다양한 잡음제거 알고리듬이 사용되고 있으나 낮은 SNR환경에서 부정확한 잡음추정으로 발생하는 스펙트럼 손상과 잔존 잡음은 음성인식기의 인식환경과 학습 환경의 불일치를 만들게 되어 인식률을 저하시키는 원인이 된다. 본 논문에서는 이와 같은 문제를 해결하기 위해 잡음제거 알고리듬으로 MMSE-STSA 방법을 사용하였고 손상된 스펙트럼을 보상하기 위해 Ideal Binary Mask를 이용하였다. 잡음환경 (SNR 15 ~ 0 dB)에 따른 실험결과 제안된 방법을 사용했을 때 향상된 스펙트럼을 얻을 수 있었고 향상된 인식성능을 확인했다.

• 한국산업정보학회논문지
• /
• 제22권4호
• /
• pp.1-9
• /
• 2017

본 논문에서는 위상 천이 디지털 홀로그래피(PSDH; Phase-Shifting Digital Holography) 및 디지털 워터마킹(Digital Watermarking) 기반 디지털 홀로그램의 이중 암호화 기술을 제안한다. 이를 위해 먼저 디지털 워터마크에 사용할 로고 영상을 정하고, 이 영상에 대한 이진 위상 컴퓨터형성 홀로그램(CGH; Computer Generated Hologram)을 반복 알고리즘을 이용하여 설계한다. 그리고 랜덤하게 발생시킨 이진 위상 마스크를 워터마크로 정하고, 설계된 이진 위상 CGH와 XOR 논리연산을 통해 워터마크 정보에 대한 키 영상을 생성한다. 그리고 물체 영상을 위상 변조하여 세기가 일정한 함수로 만든 후, 워터마크인 랜덤하게 발생시킨 이진 위상 마스크를 곱하여 물체파를 생성한다. 이 물체파는 워터마크 정보가 포함된 잡음과 유사한 패턴을 가지는 1차 암호화된 영상이라고 할 수 있다. 이를 2-단계 PSDH기술을 적용하여 기준파와 간섭을 시키면 가시성이 향상된 최종 간섭무늬를 얻는다. 이 간섭패턴이 최종적으로 구하고자 하는 물체 영상의 2차 암호화된 영상이 된다. 암호화된 영상의 복호화는 2-단계 PSDH기술을 통한 암호화된 영상들을 이용하여 적절한 산술연산 처리한 후, 프레즈넬 변환 및 1차 암호화 과정의 역순으로 진행하면 된다. 제안된 방법의 암호화 및 복호화 기술은 컴퓨터 시뮬레이션을 통하여 검증된다.

• 반도체디스플레이기술학회지
• /
• 제15권3호
• /
• pp.1-5
• /
• 2016

In order to apply sub-resolution assist feature (SRAF) in extreme ultraviolet lithography, the maximum non-printing SRAF width and lithography process margin needs to be improved. Through simulation, we confirmed that the maximum SRAF width of 6% attenuated phase shift mask (PSM) is large compared to conventional binary intensity mask. The increase in SRAF width is due to dark region's reflectivity of PSM which consequently improves the process window. Furthermore, the critical dimension error caused by variation of SRAF width and center position is reduced by lower change in diffraction amplitude. Therefore, we speculate that the margin of SRAF application will be improved by using PSM.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 ITC-CSCC -2
• /
• pp.851-854
• /
• 2000

Digital halftoning is a technique to make an equivalent binary image from scanned photo or graphic images. Low pass filtering characteristic of human visual system can be applied to get the effect of spatial averaging of local area consisted of black and white pixels for gray image. The overlapping of black dot decreases brightness and black dot is very sensitive to human visual system in the bright region. In this paper, for gray-level expression, only bright gray region in the color image is considered for blue noise mask (BNM) approach. To solve this problem, BNM with CMY dot is used for the bright region instead of black dot. Dot-on-dot model with single mask causes the problem making much black dot overlap, color distortion. Therefore approach with three masks for C, M and Y each is proposed to decrease pixel overlap and color distortion.

• 한국진공학회:학술대회논문집
• /
• 한국진공학회 2009년도 제36회 동계학술대회 초록집
• /
• pp.250.2-250.2
• /
• 2009

• 방송공학회논문지
• /
• 제27권7호
• /
• pp.999-1010
• /
• 2022

The increase in vehicle purchases worldwide is having a very significant impact on the availability of parking spaces. In particular, since it is difficult to secure a parking space in an urban area, it may be of great help to the driver to check vehicle parking information in advance. However, the current parking lot information is still operated semi-manually, such as notifications. Therefore, in this study, we propose a system for detecting a parking space using a relatively simple image processing method based on an image taken from the sky and evaluate its performance. The proposed method first converts the captured RGB image into a black-and-white binary image. This is to simplify the calculation for detection using discrete information. Next, a morphological operation is applied to increase the clarity of the binary image, and a template mask in the form of a bounding box indicating a parking space is applied to check the parking state. Twelve image samples and 2181 total of test, were used for the experiment, and a threshold of 40% was used to detect each parking space. The experimental results showed that information on the availability of parking spaces for parking users was provided with an accuracy of 95%. Although the number of experimental images is somewhat insufficient to address the generality of accuracy, it is possible to confirm the possibility of parking space detection with a simple image processing method.

• 한국컴퓨터정보학회논문지
• /
• 제4권4호
• /
• pp.34-39
• /
• 1999

본 논문에서는 이진 영상에서 경계에 발생하는 잡영을 효율적으로 제거하고 형상을 단순화시켜 윤곽선을 추출할 수 있는 방법을 제안하였다. 제안된 방법은 이진 영상에서 영역의 윤곽선을 구하는 기존의 $2{times}2$ 마스크 사용 방법을 일부 수정하여 한 픽셀 두께의 잡영을 효율적으로 제거할 수 있도록 하였다. 이를 위해 영역 경계의 잡영에서는 윤곽선 추적 경로가 중복되는 특성과 잡영의 끝점에서의 추적 특성을 분석하여 이용하였다. 또한 흰색 바탕을 윤곽선 추출에 활용함으로써 본래의 형상을 유지하며 효과적으로 단순화된 윤곽선 추출 결과를 얻을 수 있도록 하였다. 제안된 방법은 다양한 실험을 통해 그 효용성을 확인하였다.