• 한국지능시스템학회논문지
• /
• 제13권3호
• /
• pp.371-376
• /
• 2003

본 논문에서는 효과적인 정보를 표현하는 Independent Component Analysis(ICA)-factorial 표현방법을 이용하여 얼굴감정 인식을 수행한다. 얼굴감정인식은 두 단계인 특징추출 과정과 인식과정에 의해 이루어진다. 먼저 특징추출방법은 주성분 분석(Principal Component Analysis)을 이용하여 얼굴영상의 고차원 공간을 저차원 특징공간으로 변환한 후 ICA-factorial 표현방법을 통해 좀 더 효과적으로 특징벡터를 추출한다. 인식단계는 최소거리 분류방법인 유클리디안 거리에 근거한 K-Nearest Neighbor 알고리즘으로 얼굴감정을 인식한다. 6개의 기본감정(기쁨, 슬픔, 화남, 놀람, 공포, 혐오)에 대해 얼굴 감정 데이터베이스를 구축하고 실험해본 결과 기존의 방법보다 좋은 인식 성능을 얻었다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2005년도 추계종합학술대회
• /
• pp.367-370
• /
• 2005

This paper proposes a novel method for content-based watermarking based on feature points of an image. At each feature point, watermark is embedded after affine normalization according to the local characteristic scale and orientation. The characteristic scale is the scale at which the normalized scale-space representation of an image attains a maximum value, and the characteristic orientation is the angle of the principal axis of an image. By binding watermarking with the local characteristics of an image, resilience against affine transformations can be obtained. Experimental results show that the proposed method is robust against various image processing steps including affine transformations, cropping, filtering, and JPEG compression.

• Journal of Electrical Engineering and Technology
• /
• 제10권4호
• /
• pp.1851-1863
• /
• 2015

– In the paper, to handle the problem of traditional bag-of-features model ignoring the spatial relationship of local features in human action recognition, we proposed a Multiscale Spatial Position Coding under Locality Constraint method. Specifically, to describe this spatial relationship, we proposed a mixed feature combining motion feature and multi-spatial-scale configuration. To utilize temporal information between features, sub spatial-temporal-volumes are built. Next, the pooled features of sub-STVs are obtained via max-pooling method. In classification stage, the Locality-Constrained Group Sparse Representation is adopted to utilize the intrinsic group information of the sub-STV features. The experimental results on the KTH, Weizmann, and UCF sports datasets show that our action recognition system outperforms the classical local ST feature-based recognition systems published recently.

• International Journal of Advanced Culture Technology
• /
• 제9권1호
• /
• pp.218-223
• /
• 2021

Pervasive enhancement and required enforcement of the Internet of Things (IoTs) in a distributed massively multiplayer online architecture have effected in massive growth of Big-Data in terms of server over-load. There have been some previous works to overcome the overloading of server works. However, there are lack of considered methods, which is commonly applicable. Therefore, we propose a combing Sparse Auto-Encoder and Load-Balancing, which is ProGReGA for Big-Data of server loads. In the process of Sparse Auto-Encoder, when it comes to selection of the feature-pattern, the less relevant feature-pattern could be eliminated from Big-Data. In relation to Load-Balancing, the alleviated degradation of ProGReGA can take advantage of the less redundant feature-pattern. That means the most relevant of Big-Data representation can work. In the performance evaluation, we can find that the proposed method have become more approachable and stable.

• 로봇학회논문지
• /
• 제16권2호
• /
• pp.103-111
• /
• 2021

This study addresses a low-cost lidar sensor-based glass feature extraction method for an accurate map representation using statistical moments, i.e. the mean and variance. Since the low-cost lidar sensor produces range-only data without intensity and multi-echo data, there are some difficulties in detecting glass-like objects. In this study, a principle that an incidence angle of a ray emitted from the lidar with respect to a glass surface is close to zero degrees is concerned for glass detection. Besides, all sensor data are preprocessed and clustered, which is represented using statistical moments as glass feature candidates. Glass features are selected among the candidates according to several conditions based on the principle and geometric relation in the global coordinate system. The accumulated glass features are classified according to the distance, which is lastly represented on the map. Several experiments were conducted in glass environments. The results showed that the proposed method accurately extracted and represented glass windows using proper parameters. The parameters were empirically designed and carefully analyzed. In future work, we will implement and perform the conventional SLAM algorithms combined with our glass feature extraction method in glass environments.

• 한국수학교육학회지시리즈D:수학교육연구
• /
• 제16권2호
• /
• pp.137-154
• /
• 2012

Comparative studies on mathematical modeling cognition feature were carried out between 15 excellent high school third-grade science students (excellent students for short) and 15 normal ones (normal students for short) in China by utilizing protocol analysis and expert-novice comparison methods and our conclusions have been drawn as below. 1. In the style, span and method of mathematical modeling problem representation, both excellent and normal students adopted symbolic and methodological representation style. However, excellent students use mechanical representation style more often. Excellent students tend to utilize multiple-representation while normal students tend to utilize simplicity representation. Excellent students incline to make use of circular representation while normal students incline to make use of one-way representation. 2. In mathematical modeling strategy use, excellent students tend to tend to use equilibrium assumption strategy while normal students tend to use accurate assumption strategy. Excellent students tend to use sample analog construction strategy while normal students tend to use real-time generation construction strategy. Excellent students tend to use immediate self-monitoring strategy while normal students tend to use review-monitoring strategy. Excellent students tend to use theoretical deduction and intuitive judgment testing strategy while normal students tend to use data testing strategy. Excellent students tend to use assumption adjustment and modeling adjustment strategy while normal students tend to use model solving adjustment strategy. 3. In the thinking, result and efficiency of mathematical modeling, excellent students give brief oral presentations of mathematical modeling, express themselves more logically, analyze problems deeply and thoroughly, have multiple, quick and flexible thinking and the utilization of mathematical modeling method is shown by inspiring inquiry, more correct results and high thinking efficiency while normal students give complicated protocol material, express themselves illogically, analyze problems superficially and obscurely, have simple, slow and rigid thinking and the utilization of mathematical modeling method is shown by blind inquiry, more fixed and inaccurate thinking and low thinking efficiency.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권6호
• /
• pp.2709-2729
• /
• 2016

Representation based classification, kernel method and sparse representation have received much attention in the field of face recognition. In this paper, we proposed an improved kernel principal component analysis method based on sparse representation to improve the accuracy and robustness for face recognition. First, the distances between the test sample and all training samples in kernel space are estimated based on collaborative representation. Second, S training samples with the smallest distances are selected, and Kernel Principal Component Analysis (KPCA) is used to extract the features that are exploited for classification. The proposed method implements the sparse representation under ℓ₂ regularization and performs feature extraction twice to improve the robustness. Also, we investigate the relationship between the accuracy and the sparseness coefficient, the relationship between the accuracy and the dimensionality respectively. The comparative experiments are conducted on the ORL, the GT and the UMIST face database. The experimental results show that the proposed method is more effective and robust than several state-of-the-art methods including Sparse Representation based Classification (SRC), Collaborative Representation based Classification (CRC), KCRC and Two Phase Test samples Sparse Representation (TPTSR).

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1987년도 한국자동제어학술회의논문집; 한국과학기술대학, 충남; 16-17 Oct. 1987
• /
• pp.680-683
• /
• 1987

The speech signal processing in the auditory system can be analysized based on two representations : Average discharge rate and Temporal discharge pattern. But the average discharge rate representation is restricted by the narrow dynamic range because of the rate saturation and the two tone suppression phenomena, and the temporal discharge pattern representation needs a sophisticate frequency analysis and synchrony measure. In this paper, a simple representation is proposed : using a model considering the interaction of Cochlear fluid-BM movement and a haircell model, the feature of speech signals (formant frequency and pitch of vowels) is easily estimated in the Average Synchronized Rate.

• 한국게임학회 논문지
• /
• 제17권5호
• /
• pp.103-112
• /
• 2017

많은 사용자가 함께 즐기는 온라인 게임(MMOGs)에서 IoT의 확장은 서버에 엄청난 부하를 지속적으로 증가시켜, 모든 데이터들이 Big-Data화 되어가는 환경에 있다. 이에 본 논문에서는 딥러닝 기법 중에서 가장 많이 사용되는 Sparse Autoencoder와 이미 잘 알려진 부하분산 알고리즘(ProGReGA-KF)을 결합한다. 기존 알고리즘 ProGReGA-KF과 본 논문에서 제안한 알고리즘을 이동 안정성으로 비교하였고, 제안한 알고리즘이 빅-데이터 환경에서 좀 더 안정적이고 확장성이 있음 시뮬레이션을 통해 보였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권11호
• /
• pp.5427-5445
• /
• 2019

While deep neural networks have achieved remarkable performance in representation learning, a huge amount of labeled training data are usually required by supervised deep models such as convolutional neural networks. In this paper, we propose a new representation learning method, namely generative adversarial networks (GAN) based bagging deep convolutional autoencoders (GAN-BDCAE), which can map data to diverse hierarchical representations in an unsupervised fashion. To boost the size of training data, to train deep model and to aggregate diverse learning machines are the three principal avenues towards increasing the capabilities of representation learning of neural networks. We focus on combining those three techniques. To this aim, we adopt GAN for realistic unlabeled sample generation and bagging deep convolutional autoencoders (BDCAE) for robust feature learning. The proposed method improves the discriminative ability of learned feature embedding for solving subsequent pattern recognition problems. We evaluate our approach on three standard benchmarks and demonstrate the superiority of the proposed method compared to traditional unsupervised learning methods.