• 방송공학회논문지
• /
• 제24권5호
• /
• pp.735-742
• /
• 2019

본 논문에서는 스파이크그램을 기반으로 음소 인식을 위한 특성을 추출하는 방법을 제안한다. 음소 인식에 널리 사용되는 푸리에 변환 기반의 특성은 청각 기관의 동작에 부합하는 과정으로 구해지지 않으며 프레임 단위로 추출되어 높은 시간 해상도를 가지지 못한다. 따라서 음소 인식의 성능 향상을 위해 높은 시간 해상도를 가지면서 인간의 청각기관을 모델링 하는 새로운 음성 특성 추출 기술이 요구된다. 본 논문에서는 청각 기관의 특성 추출 및 전달 과정을 모델링 하는 기법인 스파이크그램을 사용하여 음성 신호를 분석하고, 이로부터 음소 인식을 위한 특성을 추출하는 방법을 제안한다. 심층 신경망 기반의 음소 인식기를 사용하여 제안한 특성의 음소 인식 성능을 측정하였고, 짧은 음소에 대해 제안 특성이 기존 푸리에 변환 기반의 특성보다 우수한 성능을 가지는 것을 확인하였다. 이 결과로부터 청각 모델을 기반으로 추출된 새로운 음성 특성을 사용하여 음소 인식이 가능함을 확인할 수 있다.

• 정보과학회 컴퓨팅의 실제 논문지
• /
• 제23권3호
• /
• pp.194-198
• /
• 2017

본 논문에서는 학술 문헌에서 표현된 단백질 간 상호 작용(Protein-Protein Interaction) 정보를 자동으로 추출하기 위한 확장된 형태의 Convolutional Neural Network (CNN) 모델을 제안한다. 이 모델은 기존에 관계 추출(Relation Extraction)을 위해 고안된 단순 자질 기반의 CNN 모델을 확장하여 다양한 전역 자질들을 추가적으로 적용함으로써 성능을 개선할 수 있는 장점이 있다. PPI 추출 성능 평가를 위해서 많이 활용되고 있는 준거 평가 컬렉션인 AIMed를 이용한 실험에서 F-스코어 기준으로 78.0%를 나타내어 현재까지 도출된 세계 최고 성능에 비해 8.3% 높은 성능을 나타내었다. 추가적으로 CNN 모델이 복잡한 언어 처리를 통한 자질 추출 작업을 하지 않고도 단백질간 상호 작용 추출에 높은 성능을 나타냄을 보였다.

• ETRI Journal
• /
• 제33권4호
• /
• pp.589-599
• /
• 2011

This paper presents an image description and matching scheme using synthetic features for a recommendation service. The recommendation service is an example of smart search because it offers something before a user's request. In the proposed extraction scheme, an image is described by synthesized spatial and statistical features. The spatial feature is designed to increase the discriminability by reflecting delicate variations. The statistical feature is designed to increase the robustness by absorbing small variations. For extracting spatial features, we partition the image into concentric circles and extract four characteristics using a spatial relation. To extract statistical features, we adapt three transforms into the image and compose a 3D histogram as the final statistical feature. The matching schemes are designed hierarchically using the proposed spatial and statistical features. The result shows that each feature is better than the compared algorithms that use spatial or statistical features. Additionally, if we adapt the proposed whole extraction and matching scheme, the overall performance will become 98.44% in terms of the correct search ratio.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2003년도 가을 학술발표논문집 Vol.30 No.2 (1)
• /
• pp.4-6
• /
• 2003

A holistic representation, such as sparse ceding or independent component analysis (ICA), was successfully applied to explain early auditory processing and sound classification. In contrast, Part-based representation is an alternative way of understanding object recognition in brain. In this paper. we employ the non-negative matrix factorization (NMF)［1］which learns parts-based representation for sound classification. Feature extraction methods from spectrogram using NMF are explained. Experimental results show that NMF-based features improve the performance of sound classification over ICA-based features.

• 전자공학회논문지B
• /
• 제33B권6호
• /
• pp.198-207
• /
• 1996

A PSRI (position, scale, and rotation invariant ) feature extraction and automatic target recognition system using a cooperative network and an MLP is proposed. We can extract position invarient features by obtaining the target center using the projection and the moment in preprocessing stage. The scale and rotation invariant features are extracted from the contour projection of the number of edge pixels on each of the concentric circles, which is input to the cooperative network. By extracting the representative PSRI features form the features and their differentiations using max-net and min-net, we can rdduce the number of input neurons of the MLP, and make the resulted automatic target recognition system less sensitive to input variances. Experiments are conduted on various complex images which are shifted, rotated, or scaled, and the results show that the proposed system is very efficient for PSRI feature extractions and automatic target recognitions.

• The Journal of the Acoustical Society of Korea
• /
• 제21권4E호
• /
• pp.156-163
• /
• 2002

We apply independent component analysis (ICA) for extracting an optimal basis to the problem of finding efficient features for representing speech signals of a given speaker The speech segments are assumed to be generated by a linear combination of the basis functions, thus the distribution of speech segments of a speaker is modeled by adapting the basis functions so that each source component is statistically independent. The learned basis functions are oriented and localized in both space and frequency, bearing a resemblance to Gabor wavelets. These features are speaker dependent characteristics and to assess their efficiency we performed speaker identification experiments and compared our results with the conventional Fourier-basis. Our results show that the proposed method is more efficient than the conventional Fourier-based features in that they can obtain a higher speaker identification rate.

• 한국멀티미디어학회논문지
• /
• 제23권8호
• /
• pp.927-939
• /
• 2020

Gait recognition can identify people's identity from a long distance, which is very important for improving the intelligence of the monitoring system. Among many human features, gait features have the advantages of being remotely available, robust, and secure. Traditional gait feature extraction, affected by the development of behavior recognition, can only rely on manual feature extraction, which cannot meet the needs of fine gait recognition. The emergence of deep convolutional neural networks has made researchers get rid of complex feature design engineering, and can automatically learn available features through data, which has been widely used. In this paper,conduct feature metric learning in the three-dimensional space by combining the three-dimensional convolution features of the gait sequence and the Siamese structure. This method can capture the information of spatial dimension and time dimension from the continuous periodic gait sequence, and further improve the accuracy and practicability of gait recognition.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1997년도 춘계학술대회 논문집
• /
• pp.455-459
• /
• 1997

A methodology which supports the feature used from design to manufacturing for sculptured pocket is newly devlored and present. The information contents in a feature can be easily conveyed from one application to another in the manufacturing domain. However, the feature generated in one application may not be directly suitable for another whitout being modified with more information. Theobjective of the paper is to parsent the methodology of decomposing a bulky feature of sculptured pocket to be removed into compact features to be efficiently machined. In particular, the paper focuses on the two task: 1) to segment horizontally a bulky feature into intermediate features by determining the adequate depth of cut and cutter size and to generate the temporal precedence graph of the intermediate features and 2)to further decompose each intermediate feature vertical into smaller manufacturing features and to apply the variable feed rate to each small feature. The proposed method will provid better efficiency in machining time and cost than the classical method which uses a long string of NC codes necessary to remove a bulky fecture.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권12호
• /
• pp.3149-3165
• /
• 2013

Content based image retrieval is increasingly gaining popularity among image repository systems as images are a big source of digital communication and information sharing. Identification of image content is done through feature extraction which is the key operation for a successful content based image retrieval system. In this paper content based image retrieval system has been developed by adopting a strategy of combining multiple features of shape, color and relevance feedback. Shape is served as a primary operation to identify images whereas color and relevance feedback have been used as supporting features to make the system more efficient and accurate. Shape features are estimated through second derivative, least square polynomial and shapes coding methods. Color is estimated through max-min mean of neighborhood intensities. A new technique has been introduced for relevance feedback without bothering the user.

• 한국음향학회지
• /
• 제21권4호
• /
• pp.156-156
• /
• 2002

We apply independent component analysis (ICA) for extracting an optimal basis to the problem of finding efficient features for representing speech signals of a given speaker The speech segments are assumed to be generated by a linear combination of the basis functions, thus the distribution of speech segments of a speaker is modeled by adapting the basis functions so that each source component is statistically independent. The learned basis functions are oriented and localized in both space and frequency, bearing a resemblance to Gabor wavelets. These features are speaker dependent characteristics and to assess their efficiency we performed speaker identification experiments and compared our results with the conventional Fourier-basis. Our results show that the proposed method is more efficient than the conventional Fourier-based features in that they can obtain a higher speaker identification rate.