摘要 :
This paper presents a theoretical comparison of early and late fusion methods. An initial discussion on the conditions to apply early or late (soft or hard) fusion is introduced. The analysis show that, if large training sets are ...
展开
This paper presents a theoretical comparison of early and late fusion methods. An initial discussion on the conditions to apply early or late (soft or hard) fusion is introduced. The analysis show that, if large training sets are available, early fusion will be the best option. If training sets are limited we must do late fusion, either soft or hard. In this latter case, the complications inherent in optimally estimating the fusion function could be avoided in exchange for lower performance. The paper also includes a comparative review of the fusion state of the art methods with the following divisions: early sensor-level fusion; early feature-level fusion; late score-level fusion (late soft fusion); and late decision-level fusion (late hard fusion). The main strengths and weaknesses of the methods are discussed.
收起
摘要 :
A plethora of image fusion algorithms have been proposed recently, yet what are optimal fusion parameters that should be used for any multi-sensor dataset cannot be defined a priori. They could be learned by evaluating all availab...
展开
A plethora of image fusion algorithms have been proposed recently, yet what are optimal fusion parameters that should be used for any multi-sensor dataset cannot be defined a priori. They could be learned by evaluating all available fusion strategies on large, representative datasets, but this is not practical and provides no guarantee that fusion performance will remain optimal should real input conditions differ from sample data. This paper proposes and examines the viability of a powerful framework for objectively optimal image fusion that explicitly optimises fusion performance for any set of input conditions. The idea is to integrate proven concepts used in objective image fusion evaluation metrics to optimally adapt the fusion process to the input conditions. Specific focus is on fusion for display, which has a broad appeal in a wide range of fusion applications as only metrics shown to be subjectively relevant are considered. The results show that the proposed framework achieves a considerable improvement in both the level and robustness of fusion performance for a wide array of multi-sensor images.
收起
摘要 :
Multi-source information fusion technologies combine multiple homogeneous or heterogeneous information sources in space dimension or time dimension according to one specific standard, and obtain consistent interpretation or descri...
展开
Multi-source information fusion technologies combine multiple homogeneous or heterogeneous information sources in space dimension or time dimension according to one specific standard, and obtain consistent interpretation or description of the measured object, then improve performance of the information system. According to the level of fusion, the fusion model usually integrates information from three levels including data level, feature level, and decision level. The system architecture used information fusion technology can be generally classified into three kinds, such as centralized fusion, distributed fusion and hybrid fusion architecture. According to the actual problem with the different information source data characteristics, we can separately adopt different level fusion methods or combine some two levels of fusion methods, and obtain better system performance.
收起
摘要 :
Information and Data Fusion is a discipline that provides methods and techniques to build Observe-Orient-Decide-Act (OODA) capabilities for various applications. There are many ways in which these methods and techniques can be cho...
展开
Information and Data Fusion is a discipline that provides methods and techniques to build Observe-Orient-Decide-Act (OODA) capabilities for various applications. There are many ways in which these methods and techniques can be chosen to provide capabilities in each phase of the OODA decision making cycle, and there are different fusion architectures, i.e., ways these methods and techniques can be applied, grouped and integrated. How one chooses the most appropriate set of methods, techniques and fusion architecture for an application depends on a number of factors. Additional factors have to be considered in the case when decision making is performed through a collaboration of a number of fusion centres on a network, defined as Distributed Data Fusion, in the context of this lecture. This lecture describes the choices for fusion architectures, the factors leading to the selection of a fusion architecture, and proposes a model to help make these choices in the case of distributed data fusion.
收起
摘要 :
Information and Data Fusion is a discipline that provides methods and techniques to build Observe-Orient-Decide-Act (OODA) capabilities for various applications. There are many ways in which these methods and techniques can be cho...
展开
Information and Data Fusion is a discipline that provides methods and techniques to build Observe-Orient-Decide-Act (OODA) capabilities for various applications. There are many ways in which these methods and techniques can be chosen to provide capabilities in each phase of the OODA decision making cycle, and there are different fusion architectures, i.e., ways these methods and techniques can be applied, grouped and integrated. How one chooses the most appropriate set of methods, techniques and fusion architecture for an application depends on a number of factors. Additional factors have to be considered in the case when decision making is performed through a collaboration of a number of fusion centres on a network, defined as Distributed Data Fusion, in the context of this lecture. This lecture describes the choices for fusion architectures, the factors leading to the selection of a fusion architecture, and proposes a model to help make these choices in the case of distributed data fusion.
收起
摘要 :
The paper presents a novel multi-fusion procedure which consists of two alternative sequences. The first sequence performs an image fusion by simultaneously using two different analytical transformations to the image sources, thus...
展开
The paper presents a novel multi-fusion procedure which consists of two alternative sequences. The first sequence performs an image fusion by simultaneously using two different analytical transformations to the image sources, thus obtaining two intermediate fused images. The second sequence combines - in base domain - this two intermediate images using a percentage involvement of the pixel value. Thus, it follows a single and final fused image which exposes a superior and perceivable information obtained throughout the proposed method, comparing to the results achieved by a contemporan used process.
收起
摘要 :
The paper presents a novel multi-fusion procedure which consists of two alternative sequences. The first sequence performs an image fusion by simultaneously using two different analytical transformations to the image sources, thus...
展开
The paper presents a novel multi-fusion procedure which consists of two alternative sequences. The first sequence performs an image fusion by simultaneously using two different analytical transformations to the image sources, thus obtaining two intermediate fused images. The second sequence combines - in base domain - this two intermediate images using a percentage involvement of the pixel value. Thus, it follows a single and final fused image which exposes a superior and perceivable information obtained throughout the proposed method, comparing to the results achieved by a contemporan used process.
收起
摘要 :
This paper describes a generic software platform which can be used for simulation, prototyping and implementation of real time data fusion systems and for training of data fusion engineers. This platform is based on the definition...
展开
This paper describes a generic software platform which can be used for simulation, prototyping and implementation of real time data fusion systems and for training of data fusion engineers. This platform is based on the definition of a set of asynchronous "data processing nodes" cooperating by interchange "lists of data". Those data processing nodes and the data they process are generic, being the data fusion engineer the responsible for implementing the actual data fusion algorithms and defining the interest pieces of data to be interchanged. This architecture has been successfully used for several real projects, such as a simulator of multiradar multitarget tracking systems, the optimization of monoradar filters, an image based tracking system, and a generic real time data fusion system for Airport surface surveillance. In the paper both the overall architecture and some of these examples are described.
收起
摘要 :
Two strategies for fusing information from multiple sources when generating predictive models in the domain of pesticide classification are investigated: i) fusing different sets of features (molecular descriptors) before building...
展开
Two strategies for fusing information from multiple sources when generating predictive models in the domain of pesticide classification are investigated: i) fusing different sets of features (molecular descriptors) before building a model and ii) fusing the classifiers built from the individual descriptor sets. An empirical investigation demonstrates that the choice of strategy can have a significant impact on the predictive performance. Furthermore, the experiment shows that the best strategy is dependent on the type of predictive model considered. When generating a decision tree for pesticide classification, a statistically significant difference in accuracy is observed in favor of combining predictions from the individual models compared to generating a single model from the fused set of molecular descriptors. On the other hand, when the model consists of an ensemble of decision trees, a statistically significant difference in accuracy is observed in favor of building the model from the fused set of descriptors compared to fusing ensemble models built from the individual sources.
收起
摘要 :
Two strategies for fusing information from multiple sources when generating predictive models in the domain of pesticide classification are investigated: i) fusing different sets of features (molecular descriptors) before building...
展开
Two strategies for fusing information from multiple sources when generating predictive models in the domain of pesticide classification are investigated: i) fusing different sets of features (molecular descriptors) before building a model and ii) fusing the classifiers built from the individual descriptor sets. An empirical investigation demonstrates that the choice of strategy can have a significant impact on the predictive performance. Furthermore, the experiment shows that the best strategy is dependent on the type of predictive model considered. When generating a decision tree for pesticide classification, a statistically significant difference in accuracy is observed in favor of combining predictions from the individual models compared to generating a single model from the fused set of molecular descriptors. On the other hand, when the model consists of an ensemble of decision trees, a statistically significant difference in accuracy is observed in favor of building the model from the fused set of descriptors compared to fusing ensemble models built from the individual sources.
收起