Earticle

In this paper, we introduce current status of the MPEG-UD and future plans for the standard.

In this paper, the requirements on MPEG User Description (MPEG-UD), and the conceptual model of MPEG-UD are introduced.

The MPEG-UD aims to achieve interoperability among different applications and services in the context. All user information, related to user’s interaction with a multimedia content or service, can be represented in the MPEG-UD format and may be safely and securely managed by the users, e.g. by separating between public and private data or different levels of private data. Some data is static in nature (e.g., user’s identity data) while other data is dynamic (e.g., user’s preferences or location). Currently, there are clear borders between static and dynamic data whereby some of them may be managed by government of state or other authorized institutions. In the future, users will be able to manage their own MPEG-UD like a certificate of authentication and provide only a selected subset of their own information to service providers. As a result, users may take advantage of MPEG-UD, e.g., in terms of more personalized services. To show the advantages of MPEG-UD in the real world, we gathered twenty-two possible use case scenarios. In this paper, we introduced the first half of defined use cases for MPEG-UD.

The MPEG-UD aims to achieve interoperability among different applications and services in the context. All user information, related to user’s interaction with a multimedia content or service, can be represented in the MPEG-UD format and may be safely and securely managed by the users, e.g. by separating between public and private data or different levels of private data. Some data is static in nature (e.g., user’s identity data) while other data is dynamic (e.g., user’s preferences or location). Currently, there are clear borders between static and dynamic data whereby some of them may be managed by government of state or other authorized institutions. In the future, users will be able to manage their own MPEG-UD like a certificate of authentication and provide only a selected subset of their own information to service providers. As a result, users may take advantage of MPEG-UD, e.g., in terms of more personalized services. To show the advantages of MPEG-UD in the real world, we gathered twenty-two possible use case scenarios. In this paper, we introduced the second half of defined use cases for MPEG-UD.

차량용 블랙박스는 사고시의 영상을 기록하고, 차량 정보 및 주행 정보 등을 저장하는 기기로서, 앞으로 점점 더 자동차 사고에 대한 결정적인 증거 효력을 발휘하게 될 것으로 보인다. 또한 전세계적인 법제화, 의무화의 흐름 속에 그 중요성은 더욱 더 강화되고 있다. 그러나 현재 널리 쓰이고 있는 블랙박스 기기들은 저장 장치에 기록된 파일들의 무결성을 보장할 수 있는 보안 기술이 거의 적용되지 않은 상태이다. 자동차 보험회사나 법정에서는 이러한 문제들이 큰 이슈가 될 수 있다. 본 논문에서는 최근 차량용 블랙박스 기술의 현황과 대 표적인 기술과 가역 워터마킹 기법을 이용하여 이를 개선하여 새롭게 디자인한 알고리즘을 제시 한다.

의료와 군사의 분야에서 원본의 이미지를 정확히 보존하면서도 압축해야 하는 필요성이 점점 커 지고 있다. 용량의 한계는 하드웨어의 발전으로 해결되고 있지만, 이미지 전송을 해야 하는 경우에는 압축을 해서 전송하는 것들이 급 변하는 세계의 변화에 필요로 대두되고 있 다. 대부분의 경우에는 HALF CONTEXT를 사용했지만 이번 논문은 FULL CONTEXT를 사용해서 이미지를 압축하려고 한다.

Testing the effectiveness of quantization matrices in (image) compression is difficult, because each image is unique and requires individualization techniques to optimize the compression ratio. However, there is usually no time to perform these kinds of techniques each time an image is compressed, and thus more general (but less effective) quantization matrices are often used. Although these matrices are standardized, there is still room for improvement. This paper proposes the usage of a popular machine learning technique, genetic algorithms, to actually perform this improvement. Although there might be some generalization issues, we believe it can be partially overcome if the right training set is chosen.

Variable Message Sign(VMS) increasing deployment as part of the ITS policy measures to provide information to road users in the area, but is not installed, the VMS status is insufficient. In this paper, in areas without such a VMS road users to road conditions and traffic information to provide Portable Variable Message Sign(PVMS) proposed. Existing VMS Unlike PVMS, Moving and Storage system configuration, operating plans, etc., should be available, so existing VMS and has the distinction of In order to operate efficiently compatible with all the power of a multi-phase transition to the proposed solar power and battery power, power systems, considering them and express terms according to the operation time to go is proposed. Is expected to be applicable based on the results of this study future battery used in a variety of portable display systems, as well as a PVMS

With the economical data service delivery, the packet transport network (PTN) technologies such as multi-protocol label switching transport profile (MPLS-TP) have emerged as an important player in the next-generation transport networks. As the PTN continues to make such considerable progress, it is being challenged by network providers who need rapid and reliable recovery capabilities to guarantee the availability of their services. In this paper, we proposes a fault detection method for point-to-point paths configurable in a PTN. Using the software defined network (SDN) architecture enforcing network agility and automation, we decrease the fault detection time between client network devices, each of which is connected to a provider network device, up to tens of milliseconds. It enables the client network devices to perform their own recovery processes within sub-50 milliseconds, so as to contribute on reducing packet loss of customer traffic.

IP Traffic is increasing dramatically, while network operators revenue is growing slowly. Therefore, most of network operators are faced with how best to upgrade their network to handle network traffic, service requirement and customer demands. Packet Optical Transport Network integrated optical layer, OTN layer and ethernet layer is a recently emerging technology as a solution to deploy cost-effective, stable and resilient network. in this paper, we describe Packet Optical Transport Network key technologies such as OTN, MPLS-TP, PBB-TE and Packet Optical Transport Network market trends. This paper suggests Packet Optical Transport Network deployment benefits and considerations in the perspective of large-scale network.

Network providers must solve many technical problems in the network operation & management aspect when they plan to transit their backbone network to the next-generation integrated transport network (like POTN). As newtorks are tranformed to the integrated transfer network, the hierarchies are simplified and the number of network elements is much smaller than legacy separated network, but the operation and management scheme is more complicated. In next generation transfer network, remote central provisioning is more important and needed new technology like SDN(Software Defined Network). In this paper, we discribe the meaning of SDN in POTN transfer network and many things in the aspect of operation and management.

With the development of modern communication technology, the desire for quality of life, to live with the needs of the monitoring of the environment is growing more and more. For this purpose, sensing devices for people’s well-being are steadily increasing, and this trend goes beyond the analysis of individual regions or studies have been extended in the direction of extraction of common environmental factors of communities. In this paper, we discuss the IoT gateway that sense the environmental information and transfer the data to the cloud or the central processing server. The characteristics of the IoT gateway are having wired or wireless interfaces and IoT interface.

The signal distortion of ECG signal due to motion artifacts has been a limitation for developing real-time wearable healthcare system. In this paper, the designed wearable platform utilizes an adaptive linear prediction filter and general filter to reduce the motion artifacts in the original ECG signal. Two filter algorithms used for baseline drift removal are evaluated to check whether real-time execution is possible on our wearable platform. The experiment results shows that the ECG signal filtered by adaptive linear prediction filter recovers from the distorted ECG signal efficiently. And we also propose an routing protocol to rapidly support the link between a fast moving wearable vital sensor node and the relay nodes in a multi-hop WSN.

소프트웨어 정의 네트워크(SDN: Software Defined Network)는 기존의 일체형 네트워크에서 제 어기능 (Control Plane)을 데이타 처리 기능 (Data Plane)으로 부터 분리함으로서 경직화된 (ossified) 네트워크 환경을 개선하고자 설계되었 다. 이는 소프트웨어 프로그래밍을 이용한 중앙 집중식 제어를 통해 네트워크 교통(Traffic) 설정과 망관리를 원활히 제공함으로서 OPEX (Operational Expenditure)와 CAPEX (Capital Expenditure)를 줄이고, 적용시간 (deployment time)을 단축할 수 있는 차세대 망기술로 각광받 고 있다. 하지만 중앙제어기의 확장성 (Scalability) 문제가 재고되지 않고는 실제 기술적용에 어려움이 있다고 여겨지고 있다. 본 논문 에서는 현재 SDN의 확장성 확보를 위한 기 술들을 분석하여 그 현황과 문제점을 파악한 후, 원활한SDN 기술 구축을 위한 진화방향들을 모 색하고 그 과제에 대하여 설명한다.

JPEG 및 디지털 영상처리 분야에서 컬러 공간 변환은 중요한 분야로서 모바일 기기에서는 연산 의 부하 구간으로 작용 가능하다. 본 논 문에서는 소프트웨어 기반 TBB (Intel Threading Builidng Blocks) 병렬 기법을 이용한 고속 컬러 영상 변환 알고리즘을 제안한다. 제안하는 소프트웨어 기반 병렬 고속화 기법의 성능 측정을 위하여 일반적인 컬러 공간 변환 연산과 TBB 병렬 기법을 사용한 알고리즘의 연산 속 도차이를 측정하였다. 실험 결과는 TBB 병렬 연산을 적용한 컬러 공간 변환이 우수한 성능 향상 을 보임으로 소프트웨어 기반 고속 병렬 처리를 통한 JPEG 및 실시간 디지털 영상처리 응용분야에 효과적으로 적용 및 모바일 기기에 적 용 가능성도 보였다.

Augmented Reality is one of the most active research area due to the increasing popularity and its hardware performance of the smart phone released in the recent years. With the development of core technologies for implementation of AR application, AR-related standard organization started the standardization of AR for the interoperability of various AR applications or services. ISO/IEC JTC1 WG1, also known as JPEG, is one of the standardization body working on AR. In this paper we look into the AR-related standard and try to find the right approach for the standardization of JPEG AR.

There are many image descriptors which represent characteristics of the images, for example, simple color histogram and complex scale invariant feature transform. One would have experiences of taking, sharing and searching pictures nowadays. Aforementioned descriptors are used when we search the picture. Novel descriptors are continuously published to deal with mobile environment which has restricted resources compared to PC’s. In this paper we overview interest point detectors and binary descriptors and look into its improvement directions.

최근 무선통신 기술의 발달 및 다양한 모바일 단말기의 확대 보급으로 사용자는 하나의 미디어 콘텐츠를 자신이 보유한 다양한 기기 및 타인의 기기에서 공유가 필요한 욕구 및 기회가 증대하고 있다. 특히 N-스크린 및 클라우드 서비스의 발전은 이러한 서비스의 확대 발전에 크게 기여하고 있으며, 현대 많은 동영상 플레이어들이 기본적으로 단일 디바이스 기반 이어보기 기능을 제공하고 있으며 N-스크 린 서비스에서는 하나의 기기에서 재생중인 콘텐츠를 중지하고 새로운 기기로 전송하여 재생하 는 미디어 동기화 기능을 제공하고 있다. 그러나 이러한 이어보기 서비스들은 동일한 플레이어 또는 서비스 제공 사업자의 전용 플레이어 사용을 전제로 하는 문제점을 안고 있다. 따라서 본 논문에서는 멀티 디바이스 및 멀티 플레이어 환경에서의 동영상 이어보기 기능을 제 공하기 위하여 서비스 구조를 제안을 목표 로 하고 있으며 이를 위하여 현재 다양한 이어보기 서비스 사례 조사를 하며 이를 통하여 최종적 으로 다양한 환경에서 효율적인 이어보 기 서비스를 가능하게 한다.

H.264/AVC is an authoritative international video coding standard, which shows more improved code and efficiency than the existing video standards. Above all, inter prediction techniques using the motion vectors of H.264/AVC did play a big role in video compression developing. After that, HEVC came out further development of the H.264/AVC. Especially, seems a lot of effort to reduce the amount of motion information, we get inspired by the benefits, algorithm using Huffman coding method is proposed to reduce motion information. Experimental results based on JM18.2 reference software was successful in bringing the compression ratio rises.

This paper proposes a hybrid algorithm by combining Lazy Snapping and Grass-fire Algorithm to make a faster and accurate 2D/3D image conversion. By removing the current noises from Optical Flow and using Lazy Snapping with Grass-fire together, the total progress will extract the object more accurately.

It is difficult to define and classify human emotion. These vague human emotion appear not in single emotion, but in combination of various emotion. This paper proposes a emotional expression algorithm using Active Appearance Model(AAM) and Fuzz k- Nearest Neighbor which give facial expression in similar with vague human emotion. Applying Mahalanobis distance on the center class, determine inclusion level between center class and each class. Also following inclusion level, appear intensity of emotion. Our emotion recognition system can recognize a complex emotion using Fuzzy k-NN classifier.

Efficient service in multimedia contents, as its size is being increased, is strongly demanded. Image signature functionality in MPEG-7 provides a way to extract image properties satisfying predefined requirements. In this paper, our objective is to evaluate that MPEG-7 image signature can be utilized for efficient and effective image indexing that provides methods for image search in massive image database. We discuss MPEG-7 image signature tools that contain methods for image signature generation. Utilizing MPEG-7 image signature, we generate image identifiers for image indexing, and evaluates them experimentally in the context of uniqueness, generation speed, necessary storage space. From the experiments, we’ve learned that MPEG-7 image signature tools generate unique index values for images, but the generation time is relatively long regarding efficient image identifier generation and the required storage space is large demanding management of variable size of identifiers.

This paper presents a method for interactive virtual manipulation of a real object using camera tracking. Our method uses a predefined 3D model of a target object for tracking and rendering. A camera pose relative to the target object is estimated using 3D model-based tracking. A texture map is extracted from the image, which the 3D model is projected with the estimated camera pose. Then a texture-mapped model is rendered with manipulations by user inputs. Experimental results demonstrate that our method provides user interactions with 3D real objects on real scenes, not augmented virtual contents.

The Park Geun-hye administration's national philosophy of "creative economy" is centered on creating jobs and providing added-value by utilizing creativity and innovation. Other countries also have various successful outcome, such as the development of industries and businesses, in which the similar policy focused on creativity was implemented in the process. Based on the recent analysis, it is expected that the ICT sector will play a critical role in supporting Korea's creative economy provided that the ICT capacity was among the highest in creativity capacities in Korea. In this paper, I propose several ICT policy recommendations that would be helpful for the Park administration's succeessful promotion on creative economy.

As industries converge with ICT (Information Communication Technologies) playing the role of a catalyst, a new paradigm of industries is emerging as a new-found concept of industries offering brand-new values and services. However, Korea is suffering from a sluggish economic growth without job growth due to its high proportion of the service sector in the entire economy and a relative slowdown in the growth of the manufacturing sector. The country, moreover, lacks a new growth engine that will give a boost to the national economy down the road. This paper attempts to predict the future economic effects and social and cultural changes that the future smart convergence era will bring to Korea in line with its definition of ‘smart convergence’ and proposes ways to develop mid- and long- term ‘smart convergence’ industries of Korea so that Korea can secure its new growth engine for its future economic development.

Large and small company, government and businesses must contract to supply goods and services in economic-ecosystem world. I suggest new notary system is required in the process of implementing the agreement and the contract. And I propose a law to improve that named ‘fair contract pubic lawyer’ can litigate about unfair trade that is find himself at notarized document with company support organization.

Recent increase in the use of big data and increasing demand for big data are required of big data strategy for activating a new policy. In this study, we analyzed the Big Data industry structure using the competitive strategy model strategies for the purpose of establishing the competitive strategy. As a result, we will have to accumulate skills focused on ICT services and manufacturing companies through increasing investment.

Recently, the big-data treatment including generation, transmitting, storing and analyzing arises as a big-issue. The phenomenon strikes a big-impact to the all environment with politics, economics, social behavior, human culture, broadcasting, education, energy, health care, amusement, et al. and makes a new paradigm 'data economics'. To accomplish a successful settling down 'the new world', the big-data sharing is necessarily needed. The sharing should be a global, so that a global platform shall be prepared. This contribution is for a successful development of a common sharing system of the Big-Data. It contains the points to be considered and a step-by-step strategy for the system on a view point of social requirement.

3GPP LTE(Long Term Evolution)/LTE-A(LTE-Advanced) are the evolution of the UMTS which are successfully commercialized in Korea and these make possible to deliver high quality multimedia services according to the users' expectations. In the more than two years since, it has been increased of a number of users and total amount of data in LTE network. And there is a limit to budget for investment of each communication company. The investment in LTE/LTE-A, it is closely related to enlarge in-building coverage. LTE/LTE-A network engineers need new Re-engineering scheme and technology to reduce cost and ensure enough in-building coverage. In this paper, we describe how to handle LTE traffic occurred in building with (Density per Sector: Non-Uniform) technique in simulation tool. And we explained on the application of different penetration loss according to the each class of building. These functions can give more reliable analysis results to increase the in-building communication quality in LTE/LTE-A.

In this paper, the simple Doppler radar of X-band having the low phase noise characteristic by using Self Oscillating Mixer (SOM) was designed. By integrating a passive planar patch antenna, the system obtain the advantages of compact sizes, low weights, and low costs. To improve phase noise of oscillator, SSPR resonator was coupled patch antenna. The phase noise is - 102.38 dBc at 100 kHz offset. which is improved than other oscillator using antenna. The output power of proposed Doppler radar was -2.21dBm and the conversion gain is 5.6dB.