• 한국문헌정보학회지
• /
• 제51권1호
• /
• pp.227-243
• /
• 2017

This study proposed a structural and semantic framework for the characterization of events and segments in Web videos that permits content-based searches and dynamic video summarization. Although MPEG-7 supports multimedia structural and semantic descriptions, it is not currently suitable for describing multimedia content on the Web. Thus, the proposed metadata framework that was designed considering Web environments provides a thorough yet simple way to describe Web video contents. Precisely, the metadata framework was constructed on the basis of Chatman's narrative theory, three multimedia metadata formats (PBCore, MPEG-7, and TV-Anytime), and social metadata. It consists of event information, eventGroup information, segment information, and video (program) information. This study also discusses how to automatically extract metadata elements including structural and semantic metadata elements from Web videos.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권1호
• /
• pp.16-37
• /
• 2022

Accurate liver segment segmentation based on radiological images is indispensable for the preoperative analysis of liver tumor resection surgery. However, most of the existing segmentation methods are not feasible to be used directly for this task due to the challenge of exact edge prediction with some tiny and slender vessels as its clinical segmentation criterion. To address this problem, we propose a novel deep learning based segmentation model, called Boundary-Aware Dual Attention Liver Segment Segmentation Model (BADA). This model can improve the segmentation accuracy of liver segments with enhancing the edges including the vessels serving as segment boundaries. In our model, the dual gated attention is proposed, which composes of a spatial attention module and a semantic attention module. The spatial attention module enhances the weights of key edge regions by concerning about the salient intensity changes, while the semantic attention amplifies the contribution of filters that can extract more discriminative feature information by weighting the significant convolution channels. Simultaneously, we build a dataset of liver segments including 59 clinic cases with dynamically contrast enhanced MRI(Magnetic Resonance Imaging) of portal vein stage, which annotated by several professional radiologists. Comparing with several state-of-the-art methods and baseline segmentation methods, we achieve the best results on this clinic liver segment segmentation dataset, where Mean Dice, Mean Sensitivity and Mean Positive Predicted Value reach 89.01%, 87.71% and 90.67%, respectively.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2001년도 하계종합학술대회 논문집(4)
• /
• pp.211-214
• /
• 2001

This paper presents an efficient fast motion estimation algorithm and image segmentation method for low bit-rate coding. First, with region split information, the algorithm splits the image having homogeneous and semantic regions like face and semantic regions in image. Then, in these regions, We find the motion vector using adaptive search window adjustment. Additionally, with this new segment based fast motion estimation, we reduce blocking artifacts by intensively coding our interesting region(face or arm) in input image. The simulation results show the improvement in coding performance and image quality.

• 로봇학회논문지
• /
• 제12권3호
• /
• pp.249-257
• /
• 2017

This paper proposes a method to segment urban scenes semantically based on location prior information. Since major scene elements in urban environments such as roads, buildings, and vehicles are often located at specific locations, using the location prior information of these elements can improve the segmentation performance. The location priors are defined in special 2D coordinates, referred to as road-normal coordinates, which are perpendicular to the orientation of the road. With the help of depth information to each element, all the possible pixels in the image are projected into these coordinates and the learned prior information is applied to those pixels. The proposed location prior can be modeled by defining a unary potential of a conditional random field (CRF) as a sum of two sub-potentials: an appearance feature-based potential and a location potential. The proposed method was validated using publicly available KITTI dataset, which has urban images and corresponding 3D depth measurements.

• 한국음향학회지
• /
• 제28권5호
• /
• pp.441-446
• /
• 2009

본 논문에서는 오디오 정보 분석을 이용하여 골프 통영상을 자동 색인하는 알고리즘을 제안한다. 제안하는 알고리즘에서는 입력되는 골프 동영상을 비디오 신호와 오디오 신호로 분리한 후에, 연속적인 오디오 스트림을 Adaboost Cascade 분류방식을 통하여 스튜디오 환경에서의 아나운서의 음성구간, 선수이름이 TV 화면에 소개 될 때 수반되는 음악구간, 선수들의 플레이에 따라 반응하는 관중들의 박수 및 환호성 소리구간, 필드에서의 레포터의 음성구간, 바다나 바람 등의 필드환경 잡음 사운드구간 등의 5가지 구간으로 분류한다. 그리고 드라이브 샷, 아이런 샷과 퍼팅 샷 시에 발생하는 스윙 사운드는 onset 검출과 변조스펙트럼 검증 방법을 통해 검출되며, 관객의 박수 소리 구간과 결합하여 액션 및 하이라이트를 효율적으로 색인할 수 있게 한다. 제안된 알고리즘은 오디오 신호의 간단한 연산을 통해 의미를 지니고 있는 기본구조들을 검출하기 때문에 골프 동영상에서 사용자가 원하는 부분을 빠르게 브라우징하는 임베이디드 시스템에 적용가능하다.

• 스마트미디어저널
• /
• 제13권2호
• /
• pp.52-61
• /
• 2024

조선소에서는 사외 적치장의 관리를 위해 일정 주기로 Unmanned Aerial Vehicle (UAV)을 이용해 항공영상을 획득하고, 이를 사람이 판독하여 적치장 현황을 파악한다. 이러한 방법은 넓은 면적의 사외 적치장 현황을 파악하는 데 상당한 시간과 인력을 요구한다. 본 논문에서는 이러한 문제점을 해결하고 정확한 사외 적치장 현황을 파악하기 위해 사전 학습된 의미론적 분할 기반 모델(Foundation Model)을 활용한 자동 관리 기술을 제안한다. 또한, 조선소 사외 적치장의 경우 관련 부품이나 장비를 포함한 공개 데이터셋이 충분하지 않기 때문에, 의미론적 분할 기반 모델에 필요한 객체 프롬프트(Prompt)를 생성하기 위한 소규모 사외 적치장 객체 데이터셋을 직접 구축하였다. 이를 이용해 객체 검출기를 소규모 데이터셋에 추가 학습하여 초기 객체 후보를 추출하고, 의미론적 분할 기반 모델인 Segment Anything Model (SAM)의 프롬프트로 활용해 정확한 의미론적 분할 결과를 얻는다. 더 나아가, 지속적인 적치장 데이터셋 수집을 위해 SAM을 활용한 훈련 데이터 생성 파이프라인을 제안한다. 본 연구에서 제안한 방법은 기존의 의미론적 분할 방법과 비교하여 평균적 4.00%p, SegFormer에 비해 5.08%p 높은 성능을 달성하였다.

• Clinical Endoscopy
• /
• 제55권3호
• /
• pp.390-400
• /
• 2022

Background/Aims: Previous artificial intelligence (AI) models attempting to segment gastric intestinal metaplasia (GIM) areas have failed to be deployed in real-time endoscopy due to their slow inference speeds. Here, we propose a new GIM segmentation AI model with inference speeds faster than 25 frames per second that maintains a high level of accuracy. Methods: Investigators from Chulalongkorn University obtained 802 histological-proven GIM images for AI model training. Four strategies were proposed to improve the model accuracy. First, transfer learning was employed to the public colon datasets. Second, an image preprocessing technique contrast-limited adaptive histogram equalization was employed to produce clearer GIM areas. Third, data augmentation was applied for a more robust model. Lastly, the bilateral segmentation network model was applied to segment GIM areas in real time. The results were analyzed using different validity values. Results: From the internal test, our AI model achieved an inference speed of 31.53 frames per second. GIM detection showed sensitivity, specificity, positive predictive, negative predictive, accuracy, and mean intersection over union in GIM segmentation values of 93%, 80%, 82%, 92%, 87%, and 57%, respectively. Conclusions: The bilateral segmentation network combined with transfer learning, contrast-limited adaptive histogram equalization, and data augmentation can provide high sensitivity and good accuracy for GIM detection and segmentation.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.221-235
• /
• 2022

Manual inspection of steel box girders on long span bridges is time-consuming and labor-intensive. The quality of inspection relies on the subjective judgements of the inspectors. This study proposes an automated approach to detect and segment cracks in high-resolution images. An end-to-end cascaded framework is proposed to first detect the existence of cracks using a deep convolutional neural network (CNN) and then segment the crack using a modified U-Net encoder-decoder architecture. A Naïve Bayes data fusion scheme is proposed to reduce the false positives and false negatives effectively. To generate the binary crack mask, first, the original images are divided into 448 × 448 overlapping image patches where these image patches are classified as cracks versus non-cracks using a deep CNN. Next, a modified U-Net is trained from scratch using only the crack patches for segmentation. A customized loss function that consists of binary cross entropy loss and the Dice loss is introduced to enhance the segmentation performance. Additionally, a Naïve Bayes fusion strategy is employed to integrate the crack score maps from different overlapping crack patches and to decide whether a pixel is crack or not. Comprehensive experiments have demonstrated that the proposed approach achieves an 81.71% mean intersection over union (mIoU) score across 5 different training/test splits, which is 7.29% higher than the baseline reference implemented with the original U-Net.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권5호
• /
• pp.2122-2140
• /
• 2020

Proper and precise embedding of commercial ads within Webpages requires Ad-hoc analysis and understanding of their content. By the successful implementation of this step, both publishers and advertisers gain mutual benefits through increasing their revenues on the one hand, and improving user experience on the other. In this research work, we propose a novel multi-level context-based ads serving approach through which ads will be served at generic publisher websites based on their contextual relevance. In the proposed approach, knowledge encoded in domain-specific and generic semantic repositories is exploited in order to analyze and segment Webpages into sets of contextually-relevant segments. Semantically-enhanced indexes are also constructed to index ads based on their textual descriptions provided by advertisers. A modified cosine similarity matching algorithm is employed to embed each ad from the Ads repository into one or more contextually-relevant segments. In order to validate our proposal, we have implemented a prototype of an ad serving system with two datasets that consist of (11429 ads and 93 documents) and (11000 documents and 15 ads), respectively. To demonstrate the effectiveness of the proposed techniques, we experimentally tested the proposed method and compared the produced results against five baseline metrics that can be used in the context of ad serving systems. In addition, we compared the results produced by our system with other state-of-the-art models. Findings demonstrate that the accuracy of conventional ad matching techniques has improved by exploiting the proposed semantically-enhanced context-based ad serving model.

• 인지과학
• /
• 제21권4호
• /
• pp.669-696
• /
• 2010

본 논문은 한 문서의 전체 의미는 각 부분의미의 합성이라는 관점에서 미리 반자동으로 구축된 감정어휘 평가사전을 기반으로 한 시스템을 제안한다. 인간의 의사 결정 과정과 유사한 방식으로 의사 결정 과정을 모델링하려는 노력으로써 본 ARSSA 시스템은 개별 리뷰의 의미값 연산과 자료 분류를 통해 감정 표현이 나타난 영화평 리뷰의 자동 등급화에 대한 연구를 수행한다. 이는 {'평점' : '리뷰'} 이항구조로 이루어진 현재의 평점 부여 형식에서 발생하는 두 변항의 불연속성 문제를 해결해보려는 목적을 가진다. 이는 어휘 의미 합성 과정에서 반영된 추상적 의미들의 합성 함수를 통해 실현될 수 있다. 시스템의 성능 실험에서 네이버 무비에서 확보한 1000개의 리뷰에 대한 10-fold 교차 검증 실험이 수행되었다. 이 실험은 기존에 부여된 평점과 비교하여 감정어휘 평가사전을 이용하였을 때 85%의 F1 Score를 보였다.