• Journal of Information Processing Systems
• /
• 제17권3호
• /
• pp.556-570
• /
• 2021

The existing video expression recognition methods mainly focus on the spatial feature extraction of video expression images, but tend to ignore the dynamic features of video sequences. To solve this problem, a multi-mode convolution neural network method is proposed to effectively improve the performance of facial expression recognition in video. Firstly, OpenFace 2.0 is used to detect face images in video, and two deep convolution neural networks are used to extract spatiotemporal expression features. Furthermore, spatial convolution neural network is used to extract the spatial information features of each static expression image, and the dynamic information feature is extracted from the optical flow information of multiple expression images based on temporal convolution neural network. Then, the spatiotemporal features learned by the two deep convolution neural networks are fused by multiplication. Finally, the fused features are input into support vector machine to realize the facial expression classification. Experimental results show that the recognition accuracy of the proposed method can reach 64.57% and 60.89%, respectively on RML and Baum-ls datasets. It is better than that of other contrast methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권10호
• /
• pp.3390-3405
• /
• 2022

Long Short-Term Memory (LSTM) combined with attention mechanism is extensively used to generate semantic sentences of images in image captioning models. However, features of salient regions and spatial information are not utilized sufficiently in most related works. Meanwhile, the LSTM also suffers from the problem of underutilized information in a single time step. In the paper, two innovative approaches are proposed to solve these problems. First, the Synergy-Gated Attention (SGA) method is proposed, which can process the spatial features and the salient region features of given images simultaneously. SGA establishes a gated mechanism through the global features to guide the interaction of information between these two features. Then, the Recurrent Fusion LSTM (RF-LSTM) mechanism is proposed, which can predict the next hidden vectors in one time step and improve linguistic coherence by fusing future information. Experimental results on the benchmark dataset of MSCOCO show that compared with the state-of-the-art methods, the proposed method can improve the performance of image captioning model, and achieve competitive performance on multiple evaluation indicators.

• Structural Monitoring and Maintenance
• /
• 제9권2호
• /
• pp.179-200
• /
• 2022

This paper proposes a data-driven methodology for online early damage identification under changing environmental conditions. The proposed method relies on two data analysis methods: feature-based method and hybrid principal component analysis (PCA) and kernel PCA to separate damage from environmental influences. First, spectral sub-band features, namely, spectral sub-band centroids (SSCs) and log spectral sub-band energies (LSSEs), are proposed as damage-sensitive features to extract damage information from measured structural responses. Second, hybrid modeling by integrating PCA and kernel PCA is performed on the spectral sub-band feature matrix for data normalization to extract both linear and nonlinear features for nonlinear procedure monitoring. After feature normalization, suppressing environmental effects, the control charts (Hotelling T² and SPE statistics) is implemented to novelty detection and distinguish damage in structures. The hybrid PCA-KPCA technique is compared to KPCA by applying support vector machine (SVM) to evaluate the effectiveness of its performance in detecting damage. The proposed method is verified through numerical and full-scale studies (a Bridge Health Monitoring (BHM) Benchmark Problem and a cable-stayed bridge in China). The results demonstrate that the proposed method can detect the structural damage accurately and reduce false alarms by suppressing the effects and interference of environmental variations.

• Communications for Statistical Applications and Methods
• /
• 제19권1호
• /
• pp.129-134
• /
• 2012

In this paper we review a bi-aspect nonparametric test for the two-sample problem under the location translation model and propose a new one to accommodate a more broad class of underlying distributions. Then we compare the performance of our proposed test with other existing ones by obtaining empirical powers through a simulation study. Then we discuss some interesting features related to the bi-aspect test with a comment on a possible expansion for the proposed test as concluding remarks.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1995년도 추계학술대회 학술발표 논문집
• /
• pp.52-58
• /
• 1995

A shared-weight neural network that performed classification based on the features extracted with the fuzzy morphological operation is introduced. Learning rules for the structuring elements, degree of membership, and weighting factors are also precisely described. In application to handwritten digit recognition problem, the fuzzy morphological shared-weight neural network produced the results which are comparable to the state-of-art for this problem.

• International Journal of Computer Science & Network Security
• /
• 제23권9호
• /
• pp.129-133
• /
• 2023

Classification or prediction problem is how to solve it using a specific feature to obtain the predicted class. A wheat seeds specifications 4 3 classes of seeds will be used in a prediction process. A multi linear regression will be built, and a prediction error ratio will be calculated. To enhance the prediction ratio an ANN model will be built and trained. The obtained results will be examined to show how to make a prediction tool capable to compute a predicted class number very close to the target class number.

• 인터넷정보학회논문지
• /
• 제21권6호
• /
• pp.13-21
• /
• 2020

안드로이드 저자 식별 연구는 좁은 범위에서는 출처를 밝히기 위한 방법으로 해석할 수 있으나, 넓은 범위에서 본다면 알려진 저작물을 통해 유사한 저작물을 식별하는 통찰력을 얻기 위한 방법으로 해석할 수 있다. 안드로이드 저자 식별 연구에서 발견되는 문제점은 안드로이드 시스템 상 중요한 코드이지만 의미가 없는 코드들로 인하여 저자의 중요한 특징을 찾기 어렵다는 것이다. 이로 인해 합법적인 코드 또는 행동들이 악성코드로 잘못 정의되기도 한다. 이를 해결하기 위하여 서바이벌 네트워크 개념을 도입하여 여러 안드로이드 앱에서 발견되는 특징들을 제거하고 저자별로 정의되는 고유한 특징들을 생존시킴으로써 문제를 해결하고자 하였다. 제안하는 프레임워크와 선행된 연구를 비교하는 실험을 진행하였으며, 440개의 저자가 식별된 앱을 대상으로 실험한 결과에서 최대 92.10%의 분류 정확도를 도출하였고 선행된 연구와 최대 3.47%의 차이를 보였다. 이는 적은 양의 학습데이터를 이용하였으나 저자별 중복된 특징 없이 고유한 특징들을 이용하였기에 선행 연구와 차이가 나타났을 것으로 해석하였다. 또한 특징 정의 방법에 따른 선행 연구와의 비교 실험에서도 적은 수의 특징으로 동일한 정확도를 보일 수 있으며, 이는 서바이벌 네트워크 개념을 통한 지속적으로 중복된 의미 없는 특징을 관리할 수 있음을 알 수 있었다.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제12권8호
• /
• pp.365-370
• /
• 2023

사람은 노화과정에 따라 발화의 호흡, 조음, 높낮이, 주파수, 언어 표현 능력 등이 변화한다. 본 논문에서는 이러한 변화로부터 발생하는 음향적, 언어적 특징을 기반으로 발화 데이터를 성인과 노인 두 그룹으로 분류하는 성능을 비교하고자 한다. 음향적 특징으로는 발화 음성의 주파수 (frequency), 진폭(amplitude), 스펙트럼(spectrum)과 관련된 특징을 사용하였으며, 언어적 특징으로는 자연어처리 분야에서 우수한 성능을 보이고 있는 한국어 대용량 코퍼스 사전학습 모델인 KoBERT를 통해 발화 전사문의 맥락 정보를 담은 은닉상태 벡터 표현을 추출하여 사용하였다. 본 논문에서는 음향적 특징과 언어적 특징을 기반으로 학습된 각 모델의 분류 성능을 확인하였다. 또한, 다운샘플링을 통해 클래스 불균형 문제를 해소한 뒤 성인과 노인 두 클래스에 대한 각 모델의 F1 점수를 확인하였다. 실험 결과로, 음향적 특징을 사용하였을 때보다 언어적 특징을 사용하였을 때 성인과 노인 분류에서 더 높은 성능을 보이는 것으로 나타났으며, 클래스 비율이 동일하더라도 노인에 대한 분류 성능보다 성인에 대한 분류 성능이 높음을 확인하였다.

• 한국CDE학회논문집
• /
• 제1권3호
• /
• pp.177-188
• /
• 1996

An approach to providing computational support with an expert shell is discussed with the scope of an industrial wire harness design, especially at a manufacturing stage. Key issues include the development of an architecture that supports a frequent design change among engineers associated with different parts of the wiring design process and the development of hierarchical representations that capture the different characteristics (e.g., connectivity, configuration) of the harnesses. The abstraction of design information results in features, while the abstraction of drawing elements leads to the definition of objects. These abstractions are essential for efficient transactions among people and computer tools in a domain that involves numerous interacting constraints. In this paper the strategy for the problem decomposition, the definition of features, and the ways in which features are shared by various operations and design changes, are discussed. We conclude with a discussion of some of the issues raised by the project and the steps underway to address them.

• 대한산업공학회지
• /
• 제20권1호
• /
• pp.99-112
• /
• 1994

This paper describe the development of a systematic method of slicing solid parts based on a data structure called Sliced Profile Data Structure(SPDS). SPDS is an augmented polygon data structure that allows multiple layers of sliced profiles to be connected together. The method consists of five steps: (1) Selection of slicing directions, (2) Determination of slicing levels, (3) Creation of sliced profiles, (4) Connection of sliced profiles, and (5) Refinement. The presented method is aimed at enhancing the applicability of CSG for manufacturing by overcoming the problem of non-uniqueness and global nature. The SPDS-based method of feature extraction is suitable for recognizing broad scope of features with detailed information. The method is also suitable for identifying the global relationships among features and is capable of incorporating the context dependency of feature extraction.