• Korean Journal of Remote Sensing
• /
• v.21 no.5
• /
• pp.405-415
• /
• 2005

It is difficult to classify tidal flat surface that is composed of a mixture of mud, sand, water and microphytobenthos. We used a Linear Spectral Unmixing (LSU) method for effectively classifying the tidal flat surface characteristics within a pixel. This study aims at 1) detecting algal mat using LSU in the Saemangeum tidal flats, 2) determining a suitable end-member selection method in tidal flats, and 3) find out a habitual characteristics of algal mat. Two types of end-member were built; one is a reference end-member derived from field spectrometer measurements and the other image end-member. A field spectrometer was used to measure spectral reflectance, and a spectral library was accomplished by shape difference of spectra, r.m.s. difference of spectra, continuum removal and Mann-Whitney U-test. Reference end-members were extracted from the spectral library. Image end-members were obtained by applying Principle Component Analysis (PCA) to an image. The LSU method was effective to detect microphytobenthos, and successfully classified the intertidal zone into algal mat, sediment, and water body components. The reference end-member was slightly more effective than the image end-member for the classification. Fine grained upper tidal flat is generally considered as a rich habitat for algal mat. We also identified unusual microphytobenthos that inhabited coarse grained lower tidal flats.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.25 no.6
• /
• pp.572-579
• /
• 2022

The analysis of foreign aircraft appearing suddenly in air defense identification zones requires a lot of cost and time. This study aims to develop a pre-trained model that can identify neighboring military aircraft based on aircraft photographs available on the web and present a model that can determine which aircraft corresponds to based on aerial photographs taken by allies. The advantages of this model are to reduce the cost and time required for model classification by proposing a pre-trained model and to improve the performance of the classifier by data augmentation of edge-detected images, cropping, flipping and so on.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.5
• /
• pp.49-54
• /
• 2022

With the recent widespread adoption of general-purpose GPUs (GPGPUs) in cloud systems, maximizing the resource utilization through multitasking in GPGPU has become an important issue. In this article, we show that resource allocation based on the workload classification of computing-bound and memory-bound is not sufficient with respect to resource utilization, and present a new thread block scheduling policy for GPGPU that makes use of fine-grained resource utilizations of each workload. Unlike previous approaches, the proposed policy reduces scheduling overhead by separating profiling and scheduling, and maximizes resource utilizations by co-locating workloads with different bottleneck resources. Through simulations under various virtual machine scenarios, we show that the proposed policy improves the GPGPU throughput by 130.6% on average and up to 161.4%.

• Journal of the Korean earth science society
• /
• v.29 no.3
• /
• pp.305-314
• /
• 2008

The purpose of this study is to analyze the concept of igneous rock textures and to uncover incorrect descriptions regarding the concept found within high school Earth Science II course seventh curriculum textbooks. Based upon this analysis suggestions will be made so as to improve descriptions regarding the concept of igneous rock texture. At least some incorrect descriptions regarding igneous rock texture were found in all the textbooks examined. Textures of volcanic rocks are described as being either fine-grained and glassy or porphyritic, while those of plutonic rocks are described as hollocrystalline, granular, coarse-grained or equigranular. These descriptions may contribute to forming and/or reinforcing misconceptions about both the classification criteria for, as well as the general concept of igneous rock textures. Therefore, some improvement schemes for the classification of igneous rock textures have been suggested. These schemes suggest that volcanic rocks be classified as either aphanitic or porphyritic, while plutonic rocks be classified as phaneritic, hollocrystalline or equigranular according to granularity, crystallinity, and both the absolute and relative sizes of the crystals within the rock.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.4 no.5
• /
• pp.859-876
• /
• 2010

In this paper, we propose a hierarchical application traffic classification system as an alternative means to overcome the limitations of the port number and payload based methodologies, which are traditionally considered traffic classification methods. The proposed system is a new classification model that hierarchically combines a binary classifier SVM and Support Vector Data Descriptions (SVDDs). The proposed system selects an optimal attribute subset from the bi-directional traffic flows generated by our traffic analysis system (KU-MON) that enables real-time collection and analysis of campus traffic. The system is composed of three layers: The first layer is a binary classifier SVM that performs rapid classification between P2P and non-P2P traffic. The second layer classifies P2P traffic into file-sharing, messenger and TV, based on three SVDDs. The third layer performs specialized classification of all individual application traffic types. Since the proposed system enables both coarse- and fine-grained classification, it can guarantee efficient resource management, such as a stable network environment, seamless bandwidth guarantee and appropriate QoS. Moreover, even when a new application emerges, it can be easily adapted for incremental updating and scaling. Only additional training for the new part of the application traffic is needed instead of retraining the entire system. The performance of the proposed system is validated via experiments which confirm that its recall and precision measures are satisfactory.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.8B
• /
• pp.921-928
• /
• 2011

The increase of Internet users and services has caused the upsurge of data traffic over the network. Nowadays, variety of Internet applications has emerged which generates complicated and diverse data traffic. For the efficient management of Internet traffic, many traffic classification methods have been proposed. But most of the methods focused on the application-level classification, not the function-level classification or state changes of applications. The functional classification of application traffic makes possible the in-detail understanding of application behavior as well as the fine-grained control of applications traffic. In this paper we proposed automata based functional classification method of IM application traffic. We verified the feasibility of the proposed method with function-level control experiment of IM application traffic.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04c
• /
• pp.519-521
• /
• 2003

개체명 인식은 효과적인 정보추출 시스템을 구축하기 위해 반드시 선행되어야 하는 작업이다. 지금까지의 개체명 인식에 관한 연구는 인명이나 조직, 장소와 같은 일반적인 개체명 인식 작업이 대부분이었다. 그러나, 효과적인 정보추출을 위해서는 이런 일반적인 개체명들을 더욱 세분화할 필요가 있다. 본 논문에서는 SVM기반 기계학습법과 기구축된 사전과의 편집거리 비교법을 이용하여 개체명을 세분화하는 방법을 제시한다. 실험은 개체명과 세분화된 범주가 부착된 공연 관련 문서 100개 중 80개는 학습집합, 20개는 실험집합으로 사용하였고 성능 평가 척도는 정확도(accuracy)를 이용해 개별적으로 평가하였다. 실험 결과 기계학습법과 사전을 이용한 방법을 결합한 모델이 가장 좋은 성능(정확도 72.91%)을 보였다.

• Language and Information
• /
• v.9 no.2
• /
• pp.23-48
• /
• 2005

This study aims to discuss the word sense distinction of Korean middle verbs for restructuring KorLexVerb 1.0. Despite the duality of its meaning and syntactic structure, the word senses of middle verb are not clearly distinguished in current dictionaries. The underspecification causes very often mismatches that a same Korean word sense is used for two different English verb senses. A close examination on the syntactic and semantic properties of middle verb shows us that the word sense distinction and the reconstruction of hierarchical structure are indispensable. Finally, by doing this fine grained word sense distinction, we propose an alternative way of classification and description of the verb polysemy for KorLexVerb 1.0 as well as for dictionary-like language resources.

• International Journal of Computer Science & Network Security
• /
• v.22 no.6
• /
• pp.97-108
• /
• 2022

The video-assisted human action recognition [1] field is one of the most active ones in computer vision research. Since the depth data [2] obtained by Kinect cameras has more benefits than traditional RGB data, research on human action detection has recently increased because of the Kinect camera. We conducted a systematic study of strategies for recognizing human activity based on deep data in this article. All methods are grouped into deep map tactics and skeleton tactics. A comparison of some of the more traditional strategies is also covered. We then examined the specifics of different depth behavior databases and provided a straightforward distinction between them. We address the advantages and disadvantages of depth and skeleton-based techniques in this discussion.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.732-734
• /
• 2023

Facial Action Units Detection (FAUs) problem focuses on identifying various detail units expressing on the human face, as defined by the Facial Action Coding System, which constitutes a fine-grained classification problem. This is a challenging task in computer vision. In this study, we propose a Prompt Tuning approach to address this problem, involving a 2-step training process. Our method demonstrates its effectiveness on the Affective in the Wild dataset, surpassing other existing methods in terms of both accuracy and efficiency.