• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2001년도 추계학술대회 학술발표 논문집
• /
• pp.60-64
• /
• 2001

In this paper, a new learning methodology for Kernel Methods is suggested that results in a sparse representation of kernel space from the training patterns for classification problems. Among the traditional algorithms of linear discriminant function(perceptron, relaxation, LMS(least mean squared), pseudoinverse), this paper shows that the relaxation procedure can obtain the maximum margin separating hyperplane of linearly separable pattern classification problem as SVM(Support Vector Machine) classifier does. The original relaxation method gives only the necessary condition of SV patterns. We suggest the sufficient condition to identify the SV patterns in the learning epochs. Experiment results show the new methods have the higher or equivalent performance compared to the conventional approach.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제14권3호
• /
• pp.162-170
• /
• 2014

Appearance-based subspace models such as eigenfaces have been widely recognized as one of the most successful approaches to face recognition and tracking. The success of eigenfaces mainly has its origins in the benefits offered by principal component analysis (PCA), the representational power of the underlying generative process for high-dimensional noisy facial image data. The sparse extension of PCA (SPCA) has recently received significant attention in the research community. SPCA functions by imposing sparseness constraints on the eigenvectors, a technique that has been shown to yield more robust solutions in many applications. However, when SPCA is applied to facial images, the time and space complexity of PCA learning becomes a critical issue (e.g., real-time tracking). In this paper, we propose a very fast and scalable greedy forward selection algorithm for SPCA. Unlike a recent semidefinite program-relaxation method that suffers from complex optimization, our approach can process several thousands of data dimensions in reasonable time with little accuracy loss. The effectiveness of our proposed method was demonstrated on real-world face recognition and tracking datasets.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권6호
• /
• pp.1946-1963
• /
• 2014

Most background subtraction methods focus on dynamic and complex scenes without considering robustness against noise. This paper proposes a background subtraction algorithm based on dictionary learning and sparse coding for handling low light conditions. The proposed method formulates background modeling as the linear and sparse combination of atoms in the dictionary. The background subtraction is considered as the difference between sparse representations of the current frame and the background model. Assuming that the projection of the noise over the dictionary is irregular and random guarantees the adaptability of the approach in large noisy scenes. Experimental results divided in simulated large noise and realistic low light conditions show the promising robustness of the proposed approach compared with other competing methods.

• 한국멀티미디어학회논문지
• /
• 제16권5호
• /
• pp.558-565
• /
• 2013

X-ray를 이용한 여성의 유방암 검사인 유방조영술은 유방암의 초기 단계에서의 진단을 위한 효과적인 방법이다. 컴퓨터 지원 검출(CAD) 시스템은 유방조영술을 통한 진단 시 의사가 놓치기 쉬운 유방암의 징후인 종괴의 검출을 도와 유방암 진단율을 높이는 수단이다. 종괴는 다양한 모양을 지니며 경계가 뚜렷하지 않기 때문에 검출이 어렵고 결과적으로 비-종괴 영역을 포함한 많은 수의 종괴 후보영역이 CAD 시스템에서 검출된다. 따라서 CAD 시스템 설계 시 검출된 많은 수의 종괴 후보영역으로부터 실제 악성 종괴 영역을 분류할 수 있도록 우수한 성능의 분류기가 요구된다. 본 논문에서는 피셔 분별 사전학습을 통해 개선된 Sparse 표현(SR) 기반 분류방법을 제안한다. 개선된 SR 기반 분류기가 기존의 CAD 시스템에서 주로 사용되어온 Support Vector Machine (SVM) 분류기 보다 우수함을 비교실험을 통해 확인했다.

• 한국정보통신학회논문지
• /
• 제18권12호
• /
• pp.2946-2952
• /
• 2014

표본 기반 초해상도(Super Resolution 이하 SR) 방법들 중 네이버 임베딩(Neighbor Embedding 이하 NE) 기법의 기본 원리는 지역적 선형 임베딩이라는 매니폴드 학습방법의 개념과 같다. 그러나, 네이버 임베딩은 국부 학습 데이터 집합의 크기가 너무 작기 때문에 이에 따른 빈약한 일반화 능력으로 인하여 알고리즘의 성능을 크게 저하시킨다. 본 논문에서는 이와 같은 문제점을 해결하기 위해서 일반화 능력이 뛰어난 Support Vector Regression(이하 SVR)을 이용한 Sparse-Neighbor 영상 표현 학습 방법에 기반한 새로운 알고리즘을 제안하였다. 저해상도 입력 영상이 주어지면 bicubic 보간법을 이용하여 확대된 영상을 얻고, 이 확대된 영상으로부터 패치를 얻은 후 저주파 패치인지 고주파 패치 인지를 판별한 후 각 영상 패치의 가중치를 얻은 후 두 개의 SVR을 훈련하였으며 훈련된 SVR을 이용하여 고해상도의 해당 화소 값을 예측하였다. 실험을 통하여 제안된 기법이 기존의 보간법 및 네이버 임베딩 기법 등에 비해 정량적인 척도 및 시각적으로 향상된 결과를 보여 주었다.

• Journal of Computing Science and Engineering
• /
• 제11권4호
• /
• pp.121-129
• /
• 2017

Feature selection has been widely established as an efficient technique for microarray data analysis. Feature selection aims to search for the most important feature/gene subset of a given dataset according to its relevance to the current target. Unsupervised feature selection is considered to be challenging due to the lack of label information. In this paper, we propose a novel method for unsupervised feature selection, which incorporates embedded learning and $l_{2,1}-norm$ sparse regression into a framework to select genes in microarray data analysis. Local tangent space alignment is applied during embedded learning to preserve the local data structure. The $l_{2,1}-norm$ sparse regression acts as a constraint to aid in learning the gene weights correlatively, by which the proposed method optimizes for selecting the informative genes which better capture the interesting natural classes of samples. We provide an effective algorithm to solve the optimization problem in our method. Finally, to validate the efficacy of the proposed method, we evaluate the proposed method on real microarray gene expression datasets. The experimental results demonstrate that the proposed method obtains quite promising performance.

• Nuclear Engineering and Technology
• /
• 제54권3호
• /
• pp.1037-1048
• /
• 2022

A radioactive isotope identification algorithm is a prerequisite for a low-resolution scintillation detector applied to an unmanned radiation monitoring system. In this paper, a sparse representation with dictionary learning approach is proposed and applied to plastic gamma-ray spectra. Label-consistent K-SVD was used to learn a discriminative dictionary for the spectra corresponding to a mixture of four isotopes (¹³³Ba, ²²Na, ¹³⁷Cs, and ⁶⁰Co). A Monte Carlo simulation was employed to produce the simulated data as learning samples. Experimental measurement was conducted to obtain practical spectra. After determining the hyper parameters, two dictionaries tailored to the learning samples were tested by varying with the source position and the measurement time. They achieved average accuracies of 97.6% and 98.0% for all testing spectra. The average accuracy of each dictionary was above 96% for spectra measured over 2 s. They also showed acceptable performance when the spectra were artificially shifted. Thus, the proposed method could be useful for identifying radioisotopes in gamma-ray spectra from a plastic scintillation detector even when a dictionary is adapted to only simulated data. Furthermore, owing to the outstanding properties of sparse representation, the proposed approach can easily be built into an insitu monitoring system.

• 디지털산업정보학회논문지
• /
• 제17권4호
• /
• pp.53-62
• /
• 2021

Deep learning models such as convolutional neural networks and recurrent neual networks process a huge amounts of data, so they require a lot of storage and consume a lot of time and power due to memory access. Recently, research is being conducted to reduce memory usage and access by compressing data using the feature that many of deep learning data are highly sparse and localized. In this paper, we propose a compression-decompression method of storing only the non-zero data and the location information of the non-zero data excluding zero data. In order to make the location information of non-zero data, the matrix data is divided into sections uniformly. And whether there is non-zero data in the corresponding section is indicated. In this case, section division is not executed only once, but repeatedly executed, and location information is stored in each step. Therefore, it can be properly compressed according to the ratio and distribution of zero data. In addition, we propose a hardware structure that enables compression and decompression without complex operations. It was designed and verified with Verilog, and it was confirmed that it can be used in hardware deep learning accelerators.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 1998년도 가을 학술발표논문집 Vol.25 No.2 (2)
• /
• pp.336-338
• /
• 1998

This paper presents a novel neural network structure to the blind deconvolution task where the input (source) to a system is not available and the source has any type of distribution including sparse distribution. We employ multiple sensors so that spatial information plays a important role. The resulting learning algorithm is linear so that it works for both sub-and super-Gaussian source. Moreover, we can successfully deconvolve the mixture of a sparse source, while most existing algorithms [5] have difficulties in this task. Computer simulations confirm the validity and high performance of the proposed algorithm.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권5호
• /
• pp.2590-2606
• /
• 2017

Generally, research of classical image classification algorithms assume that training data and testing data are derived from the same domain with the same distribution. Unfortunately, in practical applications, this assumption is rarely met. Aiming at the problem, a domain adaption image classification approach based on multi-sparse representation is proposed in this paper. The existences of intermediate domains are hypothesized between the source and target domains. And each intermediate subspace is modeled through online dictionary learning with target data updating. On the one hand, the reconstruction error of the target data is guaranteed, on the other, the transition from the source domain to the target domain is as smooth as possible. An augmented feature representation produced by invariant sparse codes across the source, intermediate and target domain dictionaries is employed for across domain recognition. Experimental results verify the effectiveness of the proposed algorithm.