• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제6권5호
• /
• pp.1400-1420
• /
• 2012

Object detection and tracking using visual sensors is a critical component of surveillance systems, which presents many challenges. This paper addresses the enhancement of object detection and tracking via the combination of multiple visual sensors. The enhancement method we introduce compensates for missed object detection based on the partial detection of objects by multiple visual sensors. When one detects an object or more visual sensors, the detected object's local positions transformed into a global object position. Local and global information exchange allows a missed local object's position to recover. However, the exchange of the information may degrade the detection and tracking performance by incorrectly recovering the local object position, which propagated by false object detection. Furthermore, local object positions corresponding to an identical object can transformed into nonequivalent global object positions because of detection uncertainty such as shadows or other artifacts. We improved the performance by preventing the propagation of false object detection. In addition, we present an evaluation method for the final global object position. The proposed method analyzed and evaluated using case studies.

• 한국콘텐츠학회:학술대회논문집
• /
• 한국콘텐츠학회 2009년도 춘계 종합학술대회 논문집
• /
• pp.23-29
• /
• 2009

Blob detection is an essential ingredient process in some computer applications such as intelligent visual surveillance. However, previous blob detection algorithms are still computationally heavy so that supporting real-time multi-channel intelligent visual surveillance in a workstation or even one-channel real-time visual surveillance in a embedded system using them turns out prohibitively difficult. In this paper, we propose a fast and precise blob detection algorithm for visual surveillance. Blob detection in visual surveillance goes through several processing steps: foreground mask extraction, foreground mask correction, and connected component labeling. Foreground mask correction necessary for a precise detection is usually accomplished using morphological operations like opening and closing. Morphological operations are computationally expensive and moreover, they are difficult to run in parallel with connected component labeling routine since they need much different processing from what connected component labeling does. In this paper, we first develop a fast and precise foreground mask correction method utilizing on neighbor pixel checking which is also employed in connected component labeling so that the developed foreground mask correction method can be incorporated into connected component labeling routine. Through experiments, it is verified that our proposed blob detection algorithm based on the foreground mask correction method developed in this paper shows better processing speed and more precise blob detection.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권9호
• /
• pp.3712-3729
• /
• 2020

Smoke detection is helpful for early fire detection. With its large coverage area and low cost, vision-based smoke detection technology is the main research direction of outdoor smoke detection. We propose a two-stage smoke detection method combined with block Deep Normalization and Convolutional Neural Network (DNCNN) and visual change image. In the first stage, each suspected smoke region is detected from each frame of the images by using block DNCNN. According to the physical characteristics of smoke diffusion, a concept of visual change image is put forward in this paper, which is constructed by the video motion change state of the suspected smoke regions, and can describe the physical diffusion characteristics of smoke in the time and space domains. In the second stage, the Support Vector Machine (SVM) classifier is used to classify the Histogram of Oriented Gradients (HOG) features of visual change images of the suspected smoke regions, in this way to reduce the false alarm caused by the smoke-like objects such as cloud and fog. Simulation experiments are carried out on two public datasets of smoke. Results show that the accuracy and recall rate of smoke detection are high, and the false alarm rate is much lower than that of other comparison methods.

• 정보처리학회 논문지
• /
• 제13권4호
• /
• pp.166-173
• /
• 2024

본 연구에서는 얼굴 동영상에서 입술의 움직임과 음성 간의 동기화 탐지 방법을 제안한다. 기존의 연구에서는 얼굴 탐지 기술로 얼굴 영역의 바운딩 박스를 도출하고, 박스의 하단 절반 영역을 시각 인코더의 입력으로 사용하여 입술-음성 동기화 탐지에 필요한 시각적인 특징을 추출하였다. 본 연구에서는 입술-음성 동기화 탐지 모델이 음성 정보의 발화 영역인 입술에 더 집중할 수 있도록 사전 학습된 시각적 Attention 기반의 인코더 도입을 제안한다. 이를 위해 음성 정보 없이 시각적 정보만으로 발화하는 말을 예측하는 독순술(Lip-Reading)에서 사용된 Visual Transformer Pooling(VTP) 모듈을 인코더로 채택했다. 그리고, 제안 방법이 학습 파라미터 수가 적음에도 불구하고 LRS2 데이터 세트에서 다섯 프레임 기준으로 94.5% 정확도를 보임으로써 최근 모델인 VocaList를 능가하는 것을 실험적으로 증명하였다. 또, 제안 방법은 학습에 사용되지 않은 Acappella 데이터셋에서도 VocaList 모델보다 8% 가량의 성능 향상이 있음을 확인하였다.

• 한국수정란이식학회지
• /
• 제31권2호
• /
• pp.137-143
• /
• 2016

This study was conducted to evaluate a visual surveillance system. The advancement of recording technology and network service make it easy to record and transfer the videos. Moreover, progressed recognition technology help to make a distinction each other. Cows show distinguishing behaviors during their estrus period. The mounting is one of the behaviors. The result was different depending on the breed of the cows and the size of the farm. In the case of Korean native cattle, the estrus detection rate was 71.15%, however, dairy cows, the estrus detection rate was 39.38%. At the farms having below 6 modules, the estrus detection rate was 87.41%. On the other hand, at the farms having over 6 modules, the estrus detection rate was 77.78%. With the proper progress, the visual surveillance system can be used to detect heat detection.

• 전자통신동향분석
• /
• 제33권4호
• /
• pp.23-32
• /
• 2018

Object detection is a challenging field in the visual understanding research area, detecting objects in visual scenes, and the location of such objects. It has recently been applied in various fields such as autonomous driving, image surveillance, and face recognition. In traditional methods of object detection, handcrafted features have been designed for overcoming various visual environments; however, they have a trade-off issue between accuracy and computational efficiency. Deep learning is a revolutionary paradigm in the machine-learning field. In addition, because deep-learning-based methods, particularly convolutional neural networks (CNNs), have outperformed conventional methods in terms of object detection, they have been studied in recent years. In this article, we provide a brief descriptive summary of several recent deep-learning methods for object detection and deep learning architectures. We also compare the performance of these methods and present a research guide of the object detection field.

• 한국산학기술학회논문지
• /
• 제21권5호
• /
• pp.630-636
• /
• 2020

본 논문에서는 군용 항공기의 시각 탐지(visual detection)를 지연시키는 위장기술에 대하여 조사하였다. 위장(camouflage)이란 관찰자에게 드러나 보이지 않도록 어떤 물체를 거짓으로 꾸미는 것으로 정의할 수 있다. 그러나 군사적 관점에서의 위장은 완전히 사라지게 하는 것이라기보다는 관찰자의 탐지시간을 연장하거나 탐지가능성(detectability)을 낮추는데 목적이 있다. 기본적으로 항공기 위장은 항공기 위치 탐지를 지연시킬 뿐만 아니라, 관측자에게 항공기의 속도와 고도, 진행방향에 대한 혼란을 유발하여야 한다. 따라서 저(低)탐지기술 또는 위장기술은 군용 항공기의 생존성 향상에 많은 영향을 미치므로 많은 연구가 지속적으로 진행되었다. 근접 지원 항공기 및 제공 전투기의 경우는 다색(multi-tone) 위장패턴과 반음영(counter-shaded) 위장패턴이 일반적으로 적용되고 있다. 아울러, 단색(mono-tone) 위장패턴 역시 색상(hue)과 명도(brightness)가 적절히 조절 및 조합되었을 때 위장효과가 큰 것으로 나타났다. 항공기의 위장 성능 향상을 위한 능동 시각 위장 기술(active camouflage techniques)에 관한 연구도 진행되었다. 특히, 발광 반사율이 높은 발광 장치를 사용하는 Counter-illumination 기술은 항공기 표면과 배경 하늘의 명도차를 최소화하여 위장 효과를 향상시켰다. 이와 같은 능동 시각 위장 기술은 시각 탐지에 비교적 취약한 저고도 무인기의 생존성 향상에 기여할 것으로 기대된다.

• 한국멀티미디어학회논문지
• /
• 제19권8호
• /
• pp.1345-1360
• /
• 2016

In this paper, we propose a new real-time human detection under omni-directional cameras for visual surveillance purpose, based on CNN with unified detection and AGMM. Compared to CNN-based state-of-the-art object detection methods. YOLO model-based object detection method boasts of very fast object detection, but with less accuracy. The proposed method adapts the unified detecting CNN of YOLO model so as to be intensified by the additional foreground contextual information obtained from pre-stage AGMM. Increased computational time incurred by additional AGMM processing is compensated by speed-up gain obtained from utilizing 2-D input data consisting of grey-level image data and foreground context information instead of 3-D color input data. Through various experiments, it is shown that the proposed method performs better with respect to accuracy and more robust to environment changes than YOLO model-based human detection method, but with the similar processing speeds to that of YOLO model-based one. Thus, it can be successfully employed for embedded surveillance application.

• 한국컴퓨터정보학회논문지
• /
• 제21권9호
• /
• pp.37-43
• /
• 2016

This paper introduces a method for video shot group detection needed for efficient management and summary of video. The proposed method detects shots based on low-level visual properties and performs temporal and spatial clustering based on visual similarity of neighboring shots. Shot groups created from temporal clustering are further clustered into small groups with respect to visual similarity. A set of representative shot frames are selected from each cluster of the smaller groups representing a scene. Shots excluded from temporal clustering are also clustered into groups from which representative shot frames are selected. A number of video clips are collected and applied to the method for accuracy of shot group detection. We achieved 91% of accuracy of the method for shot group detection. The number of representative shot frames is reduced to 1/3 of the total shot frames. The experiment also shows the inverse relationship between accuracy and compression rate.

• 한국컴퓨터정보학회논문지
• /
• 제25권7호
• /
• pp.57-64
• /
• 2020

본 논문에서는 상관필터를 이용한 영상 추적에서 탐색 영역의 크기 조절이 가능한 재탐지 방법을 제안한다. 실제 장비를 통해 영상 추적 기능을 실행할 때에는 표적이 특정 물체에 가리고 다시 나타나는 일이 빈번하게 일어나는데, 따라서 표적의 소실 판단과 재탐지 방법이 필요하다. 본 알고리즘은 강인한 추적을 위해 커널 상관필터를 사용한다. 일반적인 상관필터를 활용한 영상 추적 알고리즘에서는 표적을 탐지하는 범위가 학습된 필터의 크기에 국한된다. 하지만 표적의 가림이 오랜 시간 지속될수록 표적의 위치는 예측된 위치에서 벗어날 가능성이 커지고, 따라서 충분히 큰 범위에서 표적의 탐색이 이루어져야 한다. 제안하는 방법은 매 프레임 2%씩 탐색 범위를 넓히며 재탐지를 시도하여 성공률을 높인다. 실험은 항공에서 촬영된 4가지 영상을 활용하였고, 제안한 알고리즘은 재탐지가 어려운 데이터셋에서도 성공적인 결과를 보였다.