The semantic gap is a big challenge in image retrieval area. Previous studies in web image retrieval have mainly focused on Relevance feedback (RF) and Latent semantic indexing (LSI) to alleviate the gap. This paper p...
详细信息
The semantic gap is a big challenge in image retrieval area. Previous studies in web image retrieval have mainly focused on Relevance feedback (RF) and Latent semantic indexing (LSI) to alleviate the gap. This paper proposes an approach base on Frequent itemset mining (FIM) and Association rule (AR) techniques, which explores the semantic association rule between the two modalities that are represented by keyword and visual feature clusters. The rules are obtained offline based on the inverted file, and utilized in query process online to realize the integration of the two modalities of web images. Our approach improves the retrieval performance and is scalable well, as well as satisfies the requirement of the web users with no additional interactions. The experiments are carried out in our web image retrieval system named VAST (VisuAl & SemanTic image search), and the results show the effectiveness of the proposed approach.
A large-scale image retrieval system for the WWW, named VAST (VisuAl & SemanTic image search), is presented in this paper. Based on the existing inverted file and visual feature clusters, we form a semantic networ...
详细信息
ISBN:
(纸本)9788955191356
A large-scale image retrieval system for the WWW, named VAST (VisuAl & SemanTic image search), is presented in this paper. Based on the existing inverted file and visual feature clusters, we form a semantic network on top of the keyword association on the visual feature clusters. The system is able to automatically combine keyword and visual features for retrieval by the semantic network The combination is automatic, simple, and very fast, which is suitable for large-scale web dataset Meanwhile, the retrieval takes advantage of the semantic contents of the images in addition to the low-level features, which remarkably improves the retrieval precision. The experimental results demonstrate the superiority of the system.
Current grid information service is centralized or hierarchical and proves inefficient as grid scale rapidly increases. The introduction of P2P techniques into grids breaks an encouraging path. However, frequent join ...
详细信息
ISBN:
(纸本)9781605580777
Current grid information service is centralized or hierarchical and proves inefficient as grid scale rapidly increases. The introduction of P2P techniques into grids breaks an encouraging path. However, frequent join and departure of resource nodes require strong self-organization capacity of system to maintain their rigid structure. Moreover, arranging identifier space for P2P nodes is knotty and has great impact on system performance. If the identifier space is too large, some nodes will be overloaded. On the contraiy, small identifier space will bring the same problem as millennium bug. To address the issues, this paper proposes a scalable DHT-based (Distributed Hash Table) Information Service (DIS) for gridsystem, which organizes grid resources into a DHT ring based on VO (Virtual Organization). To save the identifier space while retaining the scalability and system performance, only stable VOs can join DIS via a new DHT node, whereas volatile VOs join DIS through being the sub-domain of other VO. Experimental results show that DIS provides rapid resource query, strong scalability and high throughput, meanwhile avoiding the key node failure as well as the bottleneck problem. Copyright 2008 ACM.
P2P based live streaming applications are growing up rapidly. However, when extending them to WLANs, the bandwidth bottleneck and the high rate of packet loss are usually the major stumbling blocks. To overcome this i...
详细信息
P2P based live streaming applications are growing up rapidly. However, when extending them to WLANs, the bandwidth bottleneck and the high rate of packet loss are usually the major stumbling blocks. To overcome this issue, a new wireless multicast agent mechanism (WiMA) based on the IP multicast scheme for buffer management and scheduling is proposed. First of all, a wireless agent selection method is presented to choose an appropriate agent for WiMA. Then this agent gains media data by interacting with wired neighbor peers as a common P2P node, and transmits these data to other wireless peers in the WLAN by means of multicast pushing and multicast patching. A normal wireless peer requests data according to the strategy of "emergent ones first". Experimental results show that the approach proposed can significantly save bandwidth in WLANs with acceptable start delay and satisfied playing continuity.
Peer-to-peer (P2P) has become an important computing model because of its adaptation, selforganization and autonomy. But how to organize dynamic nodes in P2P networks efficiently for gaming message passing services is...
详细信息
Peers in multiple P2P applications execute overlapped operations with different implementation, like neighbour management, resource publish/lookup, delay investigation, which leads to poor interoperability and unneces...
详细信息
With the development of semantic Web, search efficiency is becoming a challenging issue. To address this problem, we design an approach which mainly has two features. First, we explore potential semantic relationship ...
详细信息
With the development of semantic Web, search efficiency is becoming a challenging issue. To address this problem, we design an approach which mainly has two features. First, we explore potential semantic relationship between different objects and make full use of them to build a semantic buffer on server end, enhancing search speed. Second, we put another relatively small semantic buffer on each client so as to adapt to each individual user's interests, further improving the whole search efficiency and reducing server's load. Through testing on a practical platform, we testified that this semantic double-buffer based search approach can not only effectively reduce user's request response time, but considerably release server's load, bringing a high scalability.
Many grid middleware such as Globus Toolkit and GPE have started to support virtual machines. They implement the remote control of virtual machine by grid service and make use of virtual machinespsila function of quic...
详细信息
ISBN:
(纸本)9781424425129
Many grid middleware such as Globus Toolkit and GPE have started to support virtual machines. They implement the remote control of virtual machine by grid service and make use of virtual machinespsila function of quick deployment to construct the execution environment of grid application fast. In order to support virtual machine in grid environment, VMgrid (Virtual Machine grid) is presented. All services in VMgrid are implemented based on secure shell host protocol (SSH) while Globus Toolkitpsilas Virtual Workspace is on WSRF. This paper gives a clear description of VMgrid. Some comparisons with Globus Toolkitpsilas Virtual Workspace in performance are made in this paper. Performance results show that VMgrid virtual workspace atomic services are more effective than that of Globus.
Heterogeneous grid workflow management plays an important role in the grid interoperation. It integrates service resources from different gridsystems and packs them as atomic services into a composite service. The fo...
详细信息
ISBN:
(纸本)9781424432806;9780769534497
Heterogeneous grid workflow management plays an important role in the grid interoperation. It integrates service resources from different gridsystems and packs them as atomic services into a composite service. The focuses of the research are how to shield the difference among the heterogeneous gridservices and how to provide a flexible way to compose services. Heterogeneous grid workflow management mechanism, based on virtual service, compiles with BPEL4WS standards and shields the differences in the aspects of both organization of service information and the types of service among heterogeneous grids. The mechanism compensates for the weakness that traditional mechanism requires prior static binding in the service selection and scheduling. The proposed mechanism can provide some new functions like service backup and dynamic service selection based on QoS to improve the flexibility and stability of workflow management. In addition, in order to address the issues of tight coupling between data and service, the proposed mechanism employs virtual data space and corresponding data transferring solution.
In order to achieve low delay services in VoIP, we can use application-layer routing scheme, which is to find proper relay nodes to intermediate voice packets from the source peer to the destination peer. The difficul...
详细信息
ISBN:
(纸本)9781605581279
In order to achieve low delay services in VoIP, we can use application-layer routing scheme, which is to find proper relay nodes to intermediate voice packets from the source peer to the destination peer. The difficulty of the scheme is how to find the proper relay nodes, which we call relay problem. This paper presents a new application-layer routing scheme, called DPLM, to resolve relay problem. We introduced the mechanism in detail, and evaluated our mechanism in NS2 network simulation environment and got a high relay node hit ratio result, which confirms the feasibility of a real system. Also we make a comparison with traditional schemes, and the results show that DPLM scheme has good performances. Copyright 2008 ACM.
暂无评论