著者

Takafumi KOSHINAKA Kentaro NAGATOMO Koichi SHINODA

出版者

The Institute of Electronics, Information and Communication Engineers

雑誌

IEICE TRANSACTIONS on Information and Systems (ISSN:09168532)

巻号頁・発行日

vol.E95-D, no.10, pp.2469-2478, 2012-10-01

---

A novel online speaker clustering method based on a generative model is proposed. It employs an incremental variant of variational Bayesian learning and provides probabilistic (non-deterministic) decisions for each input utterance, on the basis of the history of preceding utterances. It can be expected to be robust against errors in cluster estimation and the classification of utterances, and hence to be applicable to many real-time applications. Experimental results show that it produces 50% fewer classification errors than does a conventional online method. They also show that it is possible to reduce the number of speech recognition errors by combining the method with unsupervised speaker adaptation.

著者

SAPAICO Luis Ricardo LAGA Hamid NAKAJIMA Masayuki

出版者

一般社団法人 電子情報通信学会

雑誌

IEICE transactions on information and systems (ISSN:09168532)

巻号頁・発行日

vol.94, no.8, pp.1671-1682, 2011-08-01

参考文献数

26

被引用文献数

1

---

We propose a system that, using video information, segments the mouth region from a face image and then detects the protrusion of the tongue from inside the oral cavity. Initially, under the assumption that the mouth is closed, we detect both mouth corners. We use a set of specifically oriented Gabor filters for enhancing horizontal features corresponding to the shadow existing between the upper and lower lips. After applying the Hough line detector, the extremes of the line that was found are regarded as the mouth corners. Detection rate for mouth corner localization is 85.33%. These points are then input to a mouth appearance model which fits a mouth contour to the image. By segmenting its bounding box we obtain a mouth template. Next, considering the symmetric nature of the mouth, we divide the template into right and left halves. Thus, our system makes use of three templates. We track the mouth in the following frames using normalized correlation for mouth template matching. Changes happening in the mouth region are directly described by the correlation value, i.e., the appearance of the tongue in the surface of the mouth will cause a decrease in the correlation coefficient through time. These coefficients are used for detecting the tongue protrusion. The right and left tongue protrusion positions will be detected by analyzing similarity changes between the right and left half-mouth templates and the currently tracked ones. Detection rates under the default parameters of our system are 90.20% for the tongue protrusion regardless of the position, and 84.78% for the right and left tongue protrusion positions. Our results demonstrate the feasibility of real-time tongue protrusion detection in vision-based systems and motivates further investigating the usage of this new modality in human-computer communication.

著者

Akira TAMAMORI Yoshihiko NANKAKU Keiichi TOKUDA

出版者

The Institute of Electronics, Information and Communication Engineers

雑誌

IEICE TRANSACTIONS on Information and Systems (ISSN:09168532)

巻号頁・発行日

vol.E95-D, no.8, pp.2074-2083, 2012-08-01

---

This paper proposes a new generative model which can deal with rotational data variations by extending Separable Lattice 2-D HMMs (SL2D-HMMs). In image recognition, geometrical variations such as size, location and rotation degrade the performance. Therefore, the appropriate normalization processes for such variations are required. SL2D-HMMs can perform an elastic matching in both horizontal and vertical directions; this makes it possible to model invariance to size and location. To deal with rotational variations, we introduce additional HMM states which represent the shifts of the state alignments among the observation lines in a particular direction. Face recognition experiments show that the proposed method improves the performance significantly for rotational variation data.

著者

Shinji KUSUMOTO Ken-ichi MATSUMOTO Tohru KIKUNO Koji TORII

出版者

The Institute of Electronics, Information and Communication Engineers

雑誌

IEICE TRANSACTIONS on Information and Systems (ISSN:09168532)

巻号頁・発行日

vol.E75-D, no.5, pp.674-680, 1992-09-25

---

This paper describes a new metric for evaluating the cost effectiveness of software reviews. The proposed metric is based on the degree to which costs needed for detecting all faults from software are reduced by software reviews in a project. The metric can be interpreted as combining two conventional metrics proposed by Fagan (1976) and Collofello and Woodfield (1989). As the proposed metric is normalized by virtual testing cost, we can compare the values of it among any different kind of projects. Using an experimental evaluation of the conventional metrics and the proposed metric for data collected in an industrial environment, we show the validity and usefulness of the proposed metric. In addition, we present a method to estimate a value of the proposed metric by using only the values which can be computed during the software reviews.

著者

Abelyn Methanie R. LAURITO Shingo TAKADA

出版者

The Institute of Electronics, Information and Communication Engineers

雑誌

IEICE TRANSACTIONS on Information and Systems (ISSN:09168532)

巻号頁・発行日

vol.E95-D, no.4, pp.1003-1011, 2012-04-01

---

The identification of functional and non-functional concerns is an important activity during requirements analysis. However, there may be conflicts between the identified concerns, and they must be discovered and resolved through trade-off analysis. Aspect-Oriented Requirements Engineering (AORE) has trade-off analysis as one of its goals, but most AORE approaches do not actually offer support for trade-off analysis; they focus on describing concerns and generating their composition. This paper proposes an approach for trade-off analysis based on AORE using use cases and the Requirements Conflict Matrix (RCM) to represent compositions. RCM shows the positive or negative effect of non-functional concerns over use cases and other non-functional concerns. Our approach is implemented within a tool called E-UCEd (Extended Use Case Editor). We also show the results of evaluating our tool.

著者

ISHII Hiroshi

出版者

一般社団法人電子情報通信学会

雑誌

IEICE transactions on information and systems (ISSN:09168532)

巻号頁・発行日

vol.87, no.6, pp.1299-1311, 2004-06-01

被引用文献数

2

---

This paper first discusses the misinterpretation of the concept of "ubiquitous computing" that Mark Weiser originally proposed in 1991. Weiser's main message was not the ubiquity of computers, but the transparency of interface that determines users' perception of digital technologies embedded in our physical environment seamlessly. To explore Weiser's philosophy of transparency in interfaces, this paper presents the design of an interface that uses glass bottles as "containers" and "controls" for digital information. The metaphor is a perfume bottle: Instead of scent, the bottles have been filled with music - classical, jazz, and techno music. Opening each bottle releases the sound of a specific instrument accompanied by dynamic colored light. Physical manipulation of the bottles - opening and closing - is the primary mode of interaction for controlling their musical contents. The bottles illustrates Mark Weiser's vision of the transparent (or invisible) interface that weaves itself into the fabric of everyday life. The bottles also exploits the emotional aspects of glass bottles that are tangible and visual, and evoke the smell of perfume and the taste of exotic beverages. This paper describes the design goals of the bottle interface, the arrangement of musical content, the implementation of the wireless electromagnetic tag technology, and the feedback from users who have played with the system.

著者

Masakatu MORII Yosuke TODO

出版者

The Institute of Electronics, Information and Communication Engineers

雑誌

IEICE TRANSACTIONS on Information and Systems (ISSN:09168532)

巻号頁・発行日

vol.E94-D, no.11, pp.2087-2094, 2011-11-01

---

In recent years, wireless LAN systems are widely used in campuses, offices, homes and so on. It is important to discuss the security aspect of wireless LAN networks in order to protect data confidentiality and integrity. The IEEE Standards Association formulated some security protocols, for example, Wired Equivalent Privacy (WEP) and Wi-Fi Protected Access Temporal Key Integrity Protocol (WPA-TKIP). However, these protocols have vulnerability for secure communication. In 2008, we proposed an efffective key recovery attack against WEP and it is called the TeAM-OK attack. In this paper, first, we present a different interpretation and the relation between other attacks and the TeAM-OK attack against WEP. Second, we present some existing attacks against WPA-TKIP and these attacks are not executable in a realistic environment. Then we propose an attack that is executable in a realistic environment against WPA-TKIP. This attack exploits the vulnerability implementation in the QoS packet processing feature of IEEE 802.11e. The receiver receives a falsification packet constructed as part of attack regardless of the setting of IEEE 802.11e. This vulnerability removes the attacker's condition that access points support IEEE 802.11e. We confirm that almost all wireless LAN implementations have this vulnerability. Therefore, almost all WPA-TKIP implementations cannot protect a system against the falsification attack in a realistic environment.

著者

Lankeshwara MUNASINGHE Ryutaro ICHISE

出版者

The Institute of Electronics, Information and Communication Engineers

雑誌

IEICE TRANSACTIONS on Information and Systems (ISSN:09168532)

巻号頁・発行日

vol.E95-D, no.3, pp.821-828, 2012-03-01

被引用文献数

30

---

Link prediction in social networks, such as friendship networks and coauthorship networks, has recently attracted a great deal of attention. There have been numerous attempts to address the problem of link prediction through diverse approaches. In the present paper, we focus on the temporal behavior of the link strength, particularly the relationship between the time stamps of interactions or links and the temporal behavior of link strength and how link strength affects future link evolution. Most previous studies have not sufficiently discussed either the impact of time stamps of the interactions or time stamps of the links on link evolution. The gap between the current time and the time stamps of the interactions or links is also important to link evolution. In the present paper, we introduce a new time-aware feature, referred to as time score, that captures the important aspects of time stamps of interactions and the temporality of the link strengths. We also analyze the effectiveness of time score with different parameter settings for different network data sets. The results of the analysis revealed that the time score was sensitive to different networks and different time measures. We applied time score to two social network data sets, namely, Facebook friendship network data set and a coauthorship network data set. The results revealed a significant improvement in predicting future links.

著者

Hikaru OOKURA Hiroshi YAMAMOTO Katsuyuki YAMAZAKI

出版者

The Institute of Electronics, Information and Communication Engineers

雑誌

IEICE TRANSACTIONS on Information and Systems (ISSN:09168532)

巻号頁・発行日

vol.E95-D, no.2, pp.546-548, 2012-02-01

---

In this paper, we have proposed a new method of observing walking traces, which can observe people's indoor movement for life-logging. Particularly emphasized new techniques in this paper are methods to detect locations, where walking directions are changed, by analyzing azimuth orientations measured by an orientation sensor of an Android mobile device, and to decide walking traces by a map matching with a vector map. The experimental evaluation has shown that the proposed method can determine the correct paths of walking traces.

著者

Jangseong KIM Joonsang BAEK Jianying ZHOU Taeshik SHON

出版者

The Institute of Electronics, Information and Communication Engineers

雑誌

IEICE TRANSACTIONS on Information and Systems (ISSN:09168532)

巻号頁・発行日

vol.E95-D, no.1, pp.117-125, 2012-01-01

---

Recently, numerous service discovery protocols have been introduced in the open literature. Unfortunately, many of them did not consider security issues, and for those that did, many security and privacy problems still remain. One important issue is to protect the privacy of a service provider while enabling an end-user to search an alternative service using multiple keywords. To deal with this issue, the existing protocols assumed that a directory server should be trusted or owned by each service provider. However, an adversary may compromise the directory server due to its openness property. In this paper, we suggest an efficient verification of service subscribers to resolve this issue and analyze its performance and security. Using this method, we propose an efficient and secure service discovery protocol protecting the privacy of a service provider while providing multiple keywords search to an end-user. Also, we provide performance and security analysis of our protocol.

著者

Min Kyoung SUNG Ki Yong LEE Jun-Bum SHIN Yon Dohn CHUNG

出版者

The Institute of Electronics, Information and Communication Engineers

雑誌

IEICE TRANSACTIONS on Information and Systems (ISSN:09168532)

巻号頁・発行日

vol.E95-D, no.1, pp.152-160, 2012-01-01

---

Recently, social network services are rapidly growing and this trend is expected to continue in the future. Social network data can be published for various purposes such as statistical analysis and population studies. When social network data are published, however, the privacy of some people may be disclosed. The most straightforward manner to preserve privacy in social network data is to remove the identifiers of persons from the social network data. However, an adversary can infer the identity of a person in the social network by using his/her background knowledge, which consists of content information such as the age, sex, or address of the person and structural information such as the number of persons having a relationship with the person. In this paper, we propose a privacy protection method for social network data. The proposed method anonymizes social network data to prevent privacy attacks that use both content and structural information, while minimizing the information loss or distortion of the anonymized social network data. Through extensive experiments, we verify the effectiveness and applicability of the proposed method.

著者

Xin HE Huiyun JING Qi HAN Xiamu NIU

出版者

The Institute of Electronics, Information and Communication Engineers

雑誌

IEICE TRANSACTIONS on Information and Systems (ISSN:09168532)

巻号頁・発行日

vol.E94-D, no.12, pp.2545-2548, 2011-12-01

---

We propose a novel saliency detection model based on Bayes' theorem. The model integrates the two parts of Bayes' equation to measure saliency, each part of which was considered separately in the previous models. The proposed model measures saliency by computing local kernel density estimation of features in the center-surround region and global kernel density estimation of features at each pixel across the whole image. Under the proposed model, a saliency detection method is presented that extracts DCT (Discrete Cosine Transform) magnitude of local region around each pixel as the feature. Experiments show that the proposed model not only performs competitively on psychological patterns and better than the current state-of-the-art models on human visual fixation data, but also is robust against signal uncertainty.

著者

Chuanjun REN Xiaomin JIA Hongbing HUANG Shiyao JIN

出版者

The Institute of Electronics, Information and Communication Engineers

雑誌

IEICE TRANSACTIONS on Information and Systems (ISSN:09168532)

巻号頁・発行日

vol.E94-D, no.11, pp.2212-2218, 2011-11-01

---

The description and analysis of emergence in complex adaptive system has recently become a topic of great interest in the field of systems, and lots of ideas and methods have been proposed. A Sign-based model of Stigmergy is proposed in this paper. Stigmergy is widely used in complex systems. We pick up “Sign” as a key notion to understand it. A definition of “Sign” is given, which reveals the Sign's nature and exploit the significations and relationships carried by the “Sign”. Then, a Sign-based model of Stigmergy is consequently developed, which captures the essential characteristics of Stigmergy. The basic architecture of Stigmergy as well as its constituents are presented and then discussed. The syntax and operational semantics of Stigmergy configurations are given. We illustrate the methodology of analyzing emergence in CAS by using our model.

著者

Nur Rohman ROSYID Masayuki OHRUI Hiroaki KIKUCHI Pitikhate SOORAKSA Masato TERADA

出版者

The Institute of Electronics, Information and Communication Engineers

雑誌

IEICE TRANSACTIONS on Information and Systems (ISSN:09168532)

巻号頁・発行日

vol.E94-D, no.11, pp.2139-2149, 2011-11-01

---

Overcoming the highly organized and coordinated malware threats by botnets on the Internet is becoming increasingly difficult. A honeypot is a powerful tool for observing and catching malware and virulent activity in Internet traffic. Because botnets use systematic attack methods, the sequences of malware downloaded by honeypots have particular forms of coordinated pattern. This paper aims to discover new frequent sequential attack patterns in malware automatically. One problem is the difficulty in identifying particular patterns from full yearlong logs because the dataset is too large for individual investigations. This paper proposes the use of a data-mining algorithm to overcome this problem. We implement the PrefixSpan algorithm to analyze malware-attack logs and then show some experimental results. Analysis of these results indicates that botnet attacks can be characterized either by the download times or by the source addresses of the bots. Finally, we use entropy analysis to reveal how frequent sequential patterns are involved in coordinated attacks.

著者

LEE Yong-Ju KIM Hag-Young LEE Cheol-Hoon

出版者

The Institute of Electronics, Information and Communication Engineers

雑誌

IEICE transactions on information and systems (ISSN:09168532)

巻号頁・発行日

vol.94, no.1, pp.60-68, 2011-01-01

参考文献数

17

被引用文献数

3

---

Infrastructure-as-a-Service (IaaS) cloud computing is emerging as a viable alternative to the acquisition and management of physical resources. The new main feature of IaaS cloud computing is the virtual machine (VM) technology which improves the flexibility of resource management. VMs use virtual machine images that are preconfigured and ready to run. Typically, VM image management uses local file copy and distribution via a network file system (NFS). This potentially means that a more efficient method can be used for VM image distribution. For efficient VM image management, we have designed and implemented a BitTorrent-based network block device (namely, BitNBD) for provisioning VM images in IaaS clouds. The BitNBD mainly provides a 'split read/write mechanism' to deal with concurrent VM instances where the same pieces of a VM are shared. With respect to the legacy BitTorrent protocol, the BitNBD enhances the piece picker policy and energy-saving mode. It is very effective in minimizing VM startup delays and providing a hibernating capability.

著者

MUNETOH Seiji NAKAMURA Megumi YOSHIHAMA Sachiko KUDO Michiharu

出版者

The Institute of Electronics, Information and Communication Engineers

雑誌

IEICE transactions on information and systems (ISSN:09168532)

巻号頁・発行日

vol.91, no.5, pp.1242-1251, 2008-05-01

被引用文献数

5

---

Computer security concerns have been rapidly increasing because of repeated security breaches and leakages of sensitive personal information. Such security breaches are mainly caused by an inappropriate management of the PCs, so maintaining integrity of the platform configuration is essential, and, verifying the integrity of the computer platform and software becomes more significant. To address these problems, the Trusted Computing Group (TCG) has developed various specifications that are used to measure the integrity of the platform based on hardware trust. In the trusted computing technology, the integrity data of each component running on the platform is recorded in the security chip and they are securely checked by a remote attestation. The infrastructure working group in the TCG is trying to define an Integrity Management Infrastructure in which the Platform Trust Services (PTS) is a new key component which deals with an Integrity Report. When we use the PTS in the target platform, it is a service component that collects and measures the runtime integrity of the target platform in a secure way. The PTS can also be used to validate the Integrity Reports. We introduce the notion of the Platform Validation Authority, a trusted third party, which verifies the composition of the integrity measurement of the target platform in the Integrity Reports. The Platform Validation Authority complements the role of the current Certificate Authority in the Public Key Infrastructure which attests to the integrity of the user identity as well as to related artifacts such as digital signatures. In this paper, we cover the research topics in this new area, the relevant technologies and open issues of the trusted computing, and the detail of our PTS implementation.

著者

KOBAYASHI Hidenori YAMASAKI Nobuyuki

出版者

一般社団法人電子情報通信学会

雑誌

IEICE transactions on information and systems (ISSN:09168532)

巻号頁・発行日

vol.86, no.10, pp.2040-2048, 2003-10-01

参考文献数

16

被引用文献数

7

---

The imprecise computation model is one of the flexible computation models used to construct real-time systems It is especially useful when the worst case execution times are difficult to estimate or the execution times vary widely. Although there are several ways to implement this model, they have not attained much attentions of real-world application programmers to date due to their unrealistic assumptions and high dependency on the execution environment In this paper, we present an integrated approach for implementing the imprecise computation model. In particular, our research covers three aspects First, we present a new imprecise computation model which consists of a mandatory part, an optional part, and another mandatory part called wind-up part This wind-up part allows application programmers to explicitly incorporate into their programs the exact operations needed for safe degradation of performance when there is a shortage in resources. Second, we describe a scheduling algorithm called Mandatory-First with Wind-up Part (M-FWP) which is based on the Earliest Deadline First strategy This algorithm, unlike scheduling algorithms developed for the classical imprecise computation model, is capable of scheduling a mandatory portion after an optional portion. Third, we present a dynamic priority server method for an efficient implementation of the M-FWP algorithm We also show that the number of the proposed server at most needed per node is one In order to estimate the performance of the proposed approach, we have implemented a real-time operating system called RT-Frontier. The experimental analyses have proven its ability to implement tasks based on the imprecise computation model without requiring any knowledge on the execution time of the optional part Moreover, it also showed performance gam over the traditional checkpointing technique.

著者

RYU Jegoon NISHIMURA Toshihiro

出版者

The Institute of Electronics, Information and Communication Engineers

雑誌

IEICE transactions on information and systems (ISSN:09168532)

巻号頁・発行日

vol.93, no.2, pp.359-366, 2010-02-01

参考文献数

30

被引用文献数

1 2

---

In this paper, Cellular Neural Networks using genetic algorithm (GA-CNNs) are designed for CMOS image noise reduction. Cellular Neural Networks (CNNs) could be an efficient way to apply to the image processing technique, since CNNs have high-speed parallel signal processing characteristics. Adaptive CNNs structure is designed for the reduction of Photon Shot Noise (PSN) changed according to the average number of photons, and the design of templates for adaptive CNNs is based on the genetic algorithm using real numbers. These templates are optimized to suppress PSN in corrupted images. The simulation results show that the adaptive GA-CNNs more efficiently reduce PSN than do the other noise reduction methods and can be used as a high-quality and low-cost noise reduction filter for PSN. The proposed method is designed for real-time implementation. Therefore, it can be used as a noise reduction filter for many commercial applications. The simulation results also show the feasibility to design the CNNs template for a variety of problems based on the statistical image model.

著者

WAKE Sanae H. ASAHI Toshiyuki

出版者

一般社団法人電子情報通信学会

雑誌

IEICE transactions on information and systems (ISSN:09168532)

巻号頁・発行日

vol.84, no.11, pp.1568-1576, 2001-11-01

被引用文献数

5

---

Our aim is to develop an intuitive and effective sound retrieval method for non-expert users. Such a retrival method should be developed to accommodate a human's perceptual features. We therefore first conducted an experiment to clarify how people represent sound. A participant listens to one sound stimulus and then conveys the sound to a partner. The results indicated that people used mostly verbal description categorized in three groups: the sound itself, the sound's situation, and the sound's impression. Based on these results, we propose three types of keywords: onomatopoeia, sound source, and adjective, which are typical keywords of the above three groups of sound description, for sound retrieval. This retrieval method was implemented for a sound database. Our method can increase the varieties of sounds able to be retrieved and allow users to intuitively search sounds because users can retrieve sounds by using keywords that are most natural to them.

著者

Yi TANG Junchen JIANG Xiaofei WANG Chengchen HU Bin LIU Zhijia CHEN

出版者

The Institute of Electronics, Information and Communication Engineers

雑誌

IEICE TRANSACTIONS on Information and Systems (ISSN:09168532)

巻号頁・発行日

vol.E93-D, no.12, pp.3232-3242, 2010-12-01

---

Multi-pattern matching is a key technique for implementing network security applications such as Network Intrusion Detection/Protection Systems (NIDS/NIPSes) where every packet is inspected against tens of thousands of predefined attack signatures written in regular expressions (regexes). To this end, Deterministic Finite Automaton (DFA) is widely used for multi-regex matching, but existing DFA-based researches have claimed high throughput at an expense of extremely high memory cost, so fail to be employed in devices such as high-speed routers and embedded systems where the available memory is quite limited. In this paper, we propose a parallel architecture of DFA called Parallel DFA (PDFA) taking advantage of the large amount of concurrent flows to increase the throughput with nearly no extra memory cost. The basic idea is to selectively store the underlying DFA in memory modules that can be accessed in parallel. To explore its potential parallelism we intensively study DFA-split schemes from both state and transition points in this paper. The performance of our approach in both the average cases and the worst cases is analyzed, optimized and evaluated by numerical results. The evaluation shows that we obtain an average speedup of 100 times compared with traditional DFA-based matching approach.