• 산업경영시스템학회지
• /
• 제38권3호
• /
• pp.56-63
• /
• 2015

Several fields of science have demanded large-scale workflow support, which requires thousands of CPU cores or more. In order to support such large-scale scientific workflows, high capacity parallel systems such as supercomputers are widely used. In order to increase the utilization of these systems, most schedulers use backfilling policy: Small jobs are moved ahead to fill in holes in the schedule when large jobs do not delay. Since an estimate of the runtime is necessary for backfilling, most parallel systems use user's estimated runtime. However, it is found to be extremely inaccurate because users overestimate their jobs. Therefore, in this paper, we propose a novel system for the runtime prediction based on workload-aware clustering with the goal of improving prediction performance. The proposed method for runtime prediction of parallel applications consists of three main phases. First, a feature selection based on factor analysis is performed to identify important input features. Then, it performs a clustering analysis of history data based on self-organizing map which is followed by hierarchical clustering for finding the clustering boundaries from the weight vectors. Finally, prediction models are constructed using support vector regression with the clustered workload data. Multiple prediction models for each clustered data pattern can reduce the error rate compared with a single model for the whole data pattern. In the experiments, we use workload logs on parallel systems (i.e., iPSC, LANL-CM5, SDSC-Par95, SDSC-Par96, and CTC-SP2) to evaluate the effectiveness of our approach. Comparing with other techniques, experimental results show that the proposed method improves the accuracy up to 69.08%.

• 한국컴퓨터정보학회논문지
• /
• 제17권12호
• /
• pp.101-109
• /
• 2012

TRUS영상에서 전립선에 대한 많은 진단과 치료 과정에서 정확한 전립선 경계의 추출이 요구된다. 여기에는 전립선 경계의 애매함, 반점, 낮은 그레이 레벨로 인하여 많은 어려움이 존재한다. 본 논문에서는 서포트 벡터와 뱀형상 윤곽선을 이용하여 TRUS영상의 자동 전립선 분할에 대한 방법을 제안한다. 이 방법은 전처리, 가버 특성 추출, 학습, 전립선 추출 단계로 구성된다. 텍스처 특성을 추출하기 위하여 가버 필터 뱅크가 사용되며, 학습 과정에서 전립선과 비전립선의 각 특성을 얻기 위하여, SVM이 사용된다. 전립선의 경계는 뱀형상 윤곽 알고리즘에 의해 추출된다. 실험 결과, 제안된 알고리즘은 인간 전문가가 추출한 경계와 비교했을 때 9.3%보다 적은 차이로 전립선 경계를 추출할 수 있었다.

• 대한전자공학회논문지SP
• /
• 제41권3호
• /
• pp.175-182
• /
• 2004

화자인증이란 생체인식 방법 중의 하나로 사람의 목소리를 이용하여 사용자를 인증하는 방법이다. 현재까지 가장 많이 사용되는 화자인증 알고리즘으로는 HMM(Hidden Markov Model)과 DTW(Dynamic Time Warping)를 들 수 있는데, 이들 알고리즘은 사용자의 등록 및 인증을 위해 많은 수의 특징벡터를 필요로 하므로 스마트 카드와 같은 메모리가 제한된 시스템에는 적용하기 어려운 단점이 있다. 본 논문에서는 SVM(Support vector Machine)을 이용함으로써 적은 양의 메모리와 적은 계산량으로 화자인증을 수행할 수 있는 방법을 제안하였으며, 이의 실시간 처리를 위해 하드웨어 구조를 제시하였다. 한국어 4연숫자 데이터베이스를 이용하여 제안한 알고리즘의 성능을 평가한 결과, 기존 알고리즘에 비해 약간의 에러율 증가가 있었으나 수행시간 및 모델크기에서는 상당한 감소를 나타내었다. SVM을 이용한 화자인증 알고리즘을 하드웨어로 구현한 결과, 소프트웨어로 구현한 경우에 비해서 훈련시간은 175분의 1, 인증시간에서는 6분의 1의 감소를 나타내었다.

• 방송공학회논문지
• /
• 제22권2호
• /
• pp.234-239
• /
• 2017

본 논문에서는 비디오 영상내 사물의 움직임의 방향과 크기의 변화를 이용한 새로운 서술자를 정의하고 이를 기반으로 하여 실시간으로 폭력 영상을 검출하는 방법을 제안한다. 새로 정의된 서술자는 폭력 행위의 움직임의 크기 및 방향 변화량이 일반적인 움직임에 비해 매우 크다는 관찰에 착안한 것이다. 일정한 프레임 동안의 서술자 값으로 이루어진 서술자 특징 벡터를 얻었고, 이것은 SVM(Support Vector Machine)으로 학습된 분류기를 통하여 폭력행위와 비폭력행위를 구별하는 데에 사용되었다. 제안하는 방법의 성능을 검증하기 위해 ViF(Violent Flow) 알고리즘과 세 종류의 데이터셋을 이용하여 비교 실험을 수행하였고, 모든 경우에서 더 우수한 성능을 보임을 확인하였다.

• 한국산학기술학회논문지
• /
• 제15권3호
• /
• pp.1691-1699
• /
• 2014

IEEE에서는 V2I, V2V 등의 무선 통신 기능을 제공하여 차량 운행의 안전을 증대 시킬 수 있는 WAVE 규격을 정의하고 있다. WAVE 규격에서는 무선 통신이 갖는 보안 취약성을 극복할 수 있도록 메시지의 암호화 및 인증 기능을 지원하고 있다. 본 논문에서는 WAVE 규격에서 지원하고 있는 보안 알고리즘들을 openssl 라이브러리와 C언어로 구현하였으며, 구현된 알고리즘들은 관련 규격들에서 제시하고 있는 테스트 벡터를 이용하여 정상 동작을 확인하고 성능을 측정하였다. 본 논문에서 구현된 보안 알고리즘들은 플랫폼에 독립적으로 구현되어, WAVE 보안 규격의 구현에 활용될 수 있을 것으로 생각된다.

• Journal of Information Processing Systems
• /
• 제16권2호
• /
• pp.329-342
• /
• 2020

In recent years, emotional text classification is one of the essential research contents in the field of natural language processing. It has been widely used in the sentiment analysis of commodities like hotels, and other commentary corpus. This paper proposes an improved W-LDA (weighted latent Dirichlet allocation) topic model to improve the shortcomings of traditional LDA topic models. In the process of the topic of word sampling and its word distribution expectation calculation of the Gibbs of the W-LDA topic model. An average weighted value is adopted to avoid topic-related words from being submerged by high-frequency words, to improve the distinction of the topic. It further integrates the highest classification of the algorithm of support vector machine based on the extracted high-quality document-topic distribution and topic-word vectors. Finally, an efficient integration method is constructed for the analysis and extraction of emotional words, topic distribution calculations, and sentiment classification. Through tests on real teaching evaluation data and test set of public comment set, the results show that the method proposed in the paper has distinct advantages compared with other two typical algorithms in terms of subject differentiation, classification precision, and F1-measure.

• 제어로봇시스템학회논문지
• /
• 제16권3호
• /
• pp.269-276
• /
• 2010

In this paper, we present the detection method of moving objects based on two background models. These background models support to understand multi layered environment belonged in images taken by shaking camera and each model is MBM(Multiple Background Model) and TMBM (Temporal Median Background Model). Because two background models are Pixel-based model, it must have noise by camera movement. Therefore correlation coefficient calculates the similarity between consecutive images and measures camera motion vector which indicates camera movement. For the calculation of correlation coefficient, we choose the selected region and searching area in the current and previous image respectively then we have a displacement vector by the correlation process. Every selected region must have its own displacement vector therefore the global maximum of a histogram of displacement vectors is the camera motion vector between consecutive images. The MBM classifies the intensity distribution of each pixel continuously related by camera motion vector to the multi clusters. However, MBM has weak sensitivity for temporal intensity variation thus we use TMBM to support the weakness of system. In the video-based experiment, we verify the presented algorithm needs around 49(ms) to generate two background models and detect moving objects.

• 센서학회지
• /
• 제29권3호
• /
• pp.194-200
• /
• 2020

Human detection is an important aspect in many video-based sensing and monitoring systems. Studies have been actively conducted for the automatic detection of humans in camera images, and various methods have been proposed. However, there are still problems in terms of performance and computational cost. In this paper, we describe a method for efficient human detection in the field of view of a camera, which may be static or moving, through multiple processing steps. A detection line is designated at the position where a human appears first in a sensing area, and only the one-dimensional gray pixel values of the line are monitored. If any noticeable change occurs in the detection line, corner detection and optical flow computation are performed in the vicinity of the detection line to confirm the change. When significant changes are observed in the corner numbers and optical flow vectors, the final determination of human presence in the monitoring area is performed using the Histograms of Oriented Gradients method and a Support Vector Machine. The proposed method requires processing only specific small areas of two consecutive gray images. Furthermore, this method enables operation not only in a static condition with a fixed camera, but also in a dynamic condition such as an operation using a camera attached to a moving vehicle.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제8권3호
• /
• pp.303-317
• /
• 2002

In this paper, we design and implement E-COIRS enabling users to query with concepts and image features used for further refining the concepts. For example, E-COIRS supports the query "retrieve images containing black home appliance to north of reception set. "The query includes two types of concepts: IS-A and composite. "home appliance"is an IS-A concept, and "reception set" is a composite concept. For evaluating such a query. E-COIRS includes three important components: a visual image indexer, thesauri and a query processor. Each pair of objects in an mage captured by the visual image indexer is converted into a triple. The triple consists of the two object identifiers (oids) and their spatial relationship. All the features of an object is referenced by its old. A composite concept is detected by the triple thesaurus and IS-A concept is recolonized by the fuzzy term thesaurus. The query processor obtains an image set by matching each triple in a user with an inverted file and CS-Tree. To support efficient storage use and fast retrieval on high-dimensional feature vectors, E-COIRS uses Cell-based Signature tree(CS-Tree). E-COIRS is a more advanced content-based image retrieval system than other systems which support only concepts or image features.

• 한국산업융합학회 논문집
• /
• 제27권1호
• /
• pp.87-95
• /
• 2024

This study is on modulo scheduling algorithms for multicore processor in machine learning applications. Machine learning algorithms are designed to perform a large amount of operations such as vectors and matrices in order to quickly process large amounts of data stream. To support such large amounts of computations, processor architectures to support applications such as artificial intelligence, neural networks, and machine learning are designed in the form of parallel processing such as multicore. To effectively utilize these multi-core hardware resources, various compiler techniques are being used and studied. In this study, among these compiler techniques, we analyzed the modular scheduler, which is especially important in one core's computation pipeline. This paper looked at and compared the iterative modular scheduler and the swing modular scheduler, which are the most widely used and studied. As a result, both schedulers provided similar performance results, and when measuring register pressure as an indicator, it was confirmed that the swing modulo scheduler provided slightly better performance. In this study, a technique that divides recurrence edge is proposed to improve the minimum initiation interval of the modulo schedulers.