• 한국지능시스템학회논문지
• /
• 제19권6호
• /
• pp.834-839
• /
• 2009

지능로봇시스템과 같은 HCI 환경에서 사람의 감정을 인식하기 위한 매개정보인 얼굴영상 기반 정서특징 검출문제는 관련분야의 매우 중요한 이슈이다. 본 논문에서는 보편화된 시스템 기반에서 임의의 사용자에 대한 정서 인식을 수행하기 위해 사람의 얼굴에서 나타나는 최적의 정서특징을 가장 효율적으로 추출하기 위한 연구로서 본 연구실에서 기존에 제안한 FACS와 AAM을 이용한 Bayesian Network 기반 얼굴표정 인식 시스템을 보완한 Advanced AAM을 기반 얼굴영상 정서 특징을 검출 시스템에 대한 연구를 진행하였다. 이를 수행하기 위하여 정규화된 이미지에서의 Statistical Shape Analysis로서 Advanced AAM과 얼굴 표정 분석 시스템인 FACS를 이용하여, 임의의 사용자에 대한 자동적인 정서특징 검출이 가능하도록 연구를 진행하였다.

• Journal of Industrial and Engineering Chemistry
• /
• 제67권
• /
• pp.380-387
• /
• 2018

Finding poly(amidoxime) (PAO) based adsorbent with better performance in U(VI) extraction from seawater is a hot research topic. By employing plasma treatment, the bi-functionalized adsorbents containing amidoxime and phosphate (labelled as $PO_4/PAO$) were successfully synthesized. The obtained $PO_4/PAO$ was characterized and applied for the potential extraction of U(VI) from aqueous solution. The results show that $-PO_4$ enhanced the hydrophilicity of PAO. $PO_4/PAO$ possesses good selective sorption ability for U(VI) and excellent reusability. The findings is helpful to understand optimizing performance of PAO based adsorbents for uranium extraction from seawater.

• Journal of Information Processing Systems
• /
• 제19권1호
• /
• pp.109-117
• /
• 2023

Defect detection is one of the key factors in fabric quality control. To improve the speed and accuracy of denim fabric defect detection, this paper proposes a defect detection algorithm based on cascading feature extraction architecture. Firstly, this paper extracts these weight parameters of the pre-trained VGG16 model on the large dataset ImageNet and uses its portability to train the defect detection classifier and the defect recognition classifier respectively. Secondly, retraining and adjusting partial weight parameters of the convolution layer were retrained and adjusted from of these two training models on the high-definition fabric defect dataset. The last step is merging these two models to get the defect detection algorithm based on cascading architecture. Then there are two comparative experiments between this improved defect detection algorithm and other feature extraction methods, such as VGG16, ResNet-50, and Xception. The results of experiments show that the defect detection accuracy of this defect detection algorithm can reach 94.3% and the speed is also increased by 1-3 percentage points.

• 정보처리학회논문지D
• /
• 제11D권1호
• /
• pp.95-104
• /
• 2004

본 논문에서는 사용자 프로파일을 기반으로 한 정보 필터링을 사용하여 학생 개인의 특성에 맞는 효율적인 원격 비디오 학습 평가 시스템을 제안한다. 비디오를 이용한 문제 출제를 위하여 위치, 크기, 그리고 컬러 정보를 기반으로 키 프레임을 추출하고 그레이 레벨 히스토그램 차이와 시간 윈도우를 이용하여 문제 출제 추간을 추출한다. 또한 효율적인 평가를 위하여 카테고리 기반 시스템과 키워드 기반 시스템을 합성하여 문제를 출제하도록 한다. 따라서 학생들은 부족한 영역을 보충하고 관심 있는 영역을 유지하면서 학업 성취도를 향상시킬 수 있다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권3호
• /
• pp.1595-1613
• /
• 2017

The IR camera and laser-based IR projector provide an effective solution for real-time collection of moving targets in RGB-D videos. Different from the traditional RGB videos, the captured depth videos are not affected by the illumination variation. In this paper, we propose a novel feature extraction framework to describe human activities based on the above optical video capturing method, namely spatial-temporal texture features for 3D human activity recognition. Spatial-temporal texture feature with depth information is insensitive to illumination and occlusions, and efficient for fine-motion description. The framework of our proposed algorithm begins with video acquisition based on laser projection, video preprocessing with visual background extraction and obtains spatial-temporal key images. Then, the texture features encoded from key images are used to generate discriminative features for human activity information. The experimental results based on the different databases and practical scenarios demonstrate the effectiveness of our proposed algorithm for the large-scale data sets.

• 방송공학회논문지
• /
• 제4권2호
• /
• pp.187-196
• /
• 1999

본 논문에서는 내용 기반 동영상 검색을 위하여 컬러 정보 및 모션 정보를 사용하는 효율적인 자동 특징 추출 알고리즘을 제안하고, 이를 동영상 검색 시스템에 적용한다. 컬러 정보의 경우 기존의 key-frame단위의 컬러 특징 추출의 한계를 극복하고, 동영상의 컬러 히스토그램 정보와 컬러의 공간분포 정보를 반영할 수 있는 컬러 특징 추출 알고리즘을 제안한다. 또한 MPEG-1 동영상 내의 모션 벡터와 컬러 정보를 조합한 컬러-모션 특징을 추출하여, 기존의 위치 기반 특징 추출 알고리즘의 한계를 극복하였다. 최종적으로 추출된 특징을 이용한 검색 시스템을 구현하여, 제안된 알고리즘의 성능을 평가하였다.

• 전자공학회논문지CI
• /
• 제38권1호
• /
• pp.46-57
• /
• 2001

시간적 특성을 가진 비디오 자료와 같은 멀티미디어의 자료에 접근하기 위해서는 내용 기반 이미지 검색 기술이 필요하다. 더욱이, 내용 기반 이미지 검색의 기본적인 기술 중의 하나가 대표키 프레임들의 추출이다. 제안된 방법을 구현했을 뿐만 아니라, 대표키 프레임들이 비디오 데이터의 특성에 대한 데이터 분석을 사용하여 추출될 수 있음을 보였다. 또한, 제안된 방법이 정확함 뿐만 아니라 효과적이라는 것을 증명하였다. 제안한 방법은 비디오 데이터 베이스를 위해 색인을 구축하는데 매우 유용하다. 그러므로 제안한 방법이 실세계에서 비디오 데이터 베이스를 구축하는데 사용되기를 기대한다.

• 제어로봇시스템학회논문지
• /
• 제15권5호
• /
• pp.488-495
• /
• 2009

This paper presents a fast feature extraction method for autonomous mobile robots utilizing parallel processing and based on OpenMP, SSE (Streaming SIMD Extension) and CUDA programming. In the first step on CPU version, the algorithms and codes are optimized and then implemented by parallel processing. The parallel algorithms are debugged to maintain the same level of performance and the process for extracting key points and obtaining dominant orientation with respect to key points is parallelized. After extraction, a parallel descriptor via SSE instructions is constructed. And the GPU version also implemented by parallel processing using CUDA based on the SIFT. The GPU-Parallel descriptor achieves an acceleration up to five times compared with the CPU-Parallel descriptor, but it shows the lower performance than CPU version. CPU version also speed-up the four and half times compared with the original SIFT while maintaining robust performance.

• Smart Structures and Systems
• /
• 제25권3호
• /
• pp.265-277
• /
• 2020

Recently, the concept of "drive-by" bridge monitoring system using indirect measurements from a passing vehicle to extract key parameters of a bridge has been rapidly developed. As one of the most key parameters of a bridge, the natural frequency has been successfully extracted theoretically and in practice using indirect measurements. The frequency of bridge is generally calculated applying Fast Fourier Transform (FFT) directly. However, it has been demonstrated that with the increase in vehicle velocity, the estimated frequency resolution of FFT will be very low causing a great extracted error. Moreover, because of the low frequency resolution, it is hard to detect the frequency drop caused by any damages or degradation of the bridge structural integrity. This paper will introduce a new technique of bridge frequency extraction based on Hilbert Transform (HT) that is not restricted to frequency resolution and can, therefore, improve identification accuracy. In this paper, deriving from the vehicle response, the closed-form solution associated with bridge frequency removing the effect of vehicle velocity is discussed in the analytical study. Then a numerical Vehicle-Bridge Interaction (VBI) model with a quarter car model is adopted to demonstrate the proposed approach. Finally, factors that affect the proposed approach are studied, including vehicle velocity, signal noise, and road roughness profile.

• IEIE Transactions on Smart Processing and Computing
• /
• 제2권5호
• /
• pp.266-276
• /
• 2013

LSB-based steganography is a simple and well known information hiding technique. In most LSB based techniques, a secret message is embedded into a specific position of LSB in the cover pixels. On the other hand, the main threat of LSB-based steganography is steganalysis. This paper proposes an asynchronous-cellular-automata(ACA)-based steganographic method, where secret bits are embedded into the selected position inside the cover pixel by ACA rule 51 and a secret key. As a result, it is very difficult for malicious users to retrieve a secret message from a cover image without knowing the secret key, even if the extraction algorithm is known. In addition, another layer of security is provided by almost random (rule-based) selection of a cover pixel for embedding using ACA and a different secret key. Finally, the experimental results show that the proposed method can be secured against the well-known steganalysis RS-attack.