This paper proposes an indexing scheme based on t- mixture model and ICA, which is more robust than Gaussian mixture modeling when atypical points (or outliers) exist or the set of data has heavy tail. This indexing s...
详细信息
This paper proposes an indexing scheme based on t- mixture model and ICA, which is more robust than Gaussian mixture modeling when atypical points (or outliers) exist or the set of data has heavy tail. This indexing scheme combines optimized vector quantizer and probabilistic approximate-based indexing scheme. Experimental results on large-scale graph database show a notable efficiency improvement with optimistic precision.
Video abstraction plays an important role in video browsing, video indexing, video retrieval and other video applications. In the paper, an automatic video abstraction method is developed based on relational graph rep...
详细信息
Video abstraction plays an important role in video browsing, video indexing, video retrieval and other video applications. In the paper, an automatic video abstraction method is developed based on relational graph representations. Firstly, a relational graph is constructed for all the frames in a video sequence. Secondly, the graph is partitioned into different connected subgraphs. Thirdly, Isomap is performed to reduce the dimensionality of the data set, and the output of Isomap is used as feature vector of the video frames. Lastly, a mixture model with model selection is introduced to generate the fine grain video abstraction with the centre of each clustering as the keyframes. Experiments are conducted on real world video sets with satisfying video abstraction results.
Computational cognitive modeling has recently emerged as one of the hottest issues in the AI area. Both symbolic approaches and connectionist approaches present their merits and demerits. Although Bayesian method is s...
详细信息
Computational cognitive modeling has recently emerged as one of the hottest issues in the AI area. Both symbolic approaches and connectionist approaches present their merits and demerits. Although Bayesian method is suggested to incorporate advantages of the two kinds of approaches above, there is no feasible Bayesian computational model concerning the entire cognitive process by now. In this paper, we propose a variation of traditional Bayesian network, namely Globally Connected and Locally Autonomic Bayesian Network (GCLABN), to formally describe a plausible cognitive model. The model adopts a unique knowledge representation strategy, which enables it to encode both symbolic concepts and their relationships within a graphical structure, and to generate cognition via a dynamic oscillating process rather than a straightforward reasoning process like traditional approaches. Then a simple simulation is employed to illustrate the properties and dynamic behaviors of the model. All these traits of the model are coincident with the recently discovered properties of the human cognitive process.
The effectiveness of ad-hoc retrieval systems appears to have reached a *** the research community is investigating an assortment of variables that will help them develop increasingly accurate retrieval *** paper pres...
详细信息
ISBN:
(纸本)9780972147903
The effectiveness of ad-hoc retrieval systems appears to have reached a *** the research community is investigating an assortment of variables that will help them develop increasingly accurate retrieval *** paper presents a new method of high accuracy retrieval of document,which is based on quotient *** describes the problem space from different granularity:sentence,paragraph and document,and studies the relations among different granularities.
The DCT-based real discrete Gabor transform (RDGT) is briefly reviewed in this paper, and then, block time-recursive algorithms for the efficient and fast computation of the RDGT coefficients and for the fast reconstr...
详细信息
The DCT-based real discrete Gabor transform (RDGT) is briefly reviewed in this paper, and then, block time-recursive algorithms for the efficient and fast computation of the RDGT coefficients and for the fast reconstruction of the original signal from the coefficients are developed in both the critical sampling case and the oversampling case. Unified parallel lattice structures for the implementation of the algorithms are studied. Computational complexity analysis and comparison have shown that the proposed algorithms provide a more efficient and faster approach for discrete Gabor transforms as compared to the existing discrete Gabor transform algorithms.
The different approaches used for target decomposition (TD) theory in radar polarimetry are reviewed and three main types of theorems are introduced: those based on Mueller matrix, those using an eigenvector analysis ...
详细信息
The different approaches used for target decomposition (TD) theory in radar polarimetry are reviewed and three main types of theorems are introduced: those based on Mueller matrix, those using an eigenvector analysis of the coherency matrix, and those employing coherent decomposition of the scattering matrix. Support vector machine (SVM), as a novel approach in pattern recognition, has demonstrated success in many fields. A new algorithm of target classification, by combining target decomposition and the support vector machine, is *** conduct the experiment, the polarimetric synthetic aperture radar (SAR) data are used. Experimental results show that it is feasible and efficient to target classification by applying target decomposition to extract scattering mechanisms, and the effects of kernel function and its parameters on the classification efficiency are significant.
The spacing distance of amino acids is used to study the relationship between the primary structure and structural classification of large proteins. Rescaled-range (R/S) analysis was adopted to examine long-range corr...
详细信息
Structure alignment could help to find shape similarities between proteins and guide structure classification and fold recognition. Common substructure detection and extraction are especially important, for which coul...
详细信息
ISBN:
(纸本)9781424415786
Structure alignment could help to find shape similarities between proteins and guide structure classification and fold recognition. Common substructure detection and extraction are especially important, for which could guide the biologist to discover binding site or active site. We represent each segment of alpha-carbon backbone by using dihedral angles and curve moment invariants. Then, local and global structure alignment could be performed by iterative closest point algorithm. Maximum common substructures between a pair of proteins or within a protein could be found. Active sites also could be detected by the proposed algorithm.
Learning methods of constructive neural network aims to overcome the disadvantages of BP algorithm, which has many advantages such as fast convergent rate, less computation, good fault-tolerant and strong generalizati...
详细信息
ISBN:
(纸本)9781615677214
Learning methods of constructive neural network aims to overcome the disadvantages of BP algorithm, which has many advantages such as fast convergent rate, less computation, good fault-tolerant and strong generalization ability, etc. Its structure is constructed step by step in the processing of the data rather than being prescribed in advance. This paper mainly introduces the developing motivations, research status quo and development directions, laying stress on two kinds of constructive learning methods: FP algorithm and covering algorithm. After discussing the construction and basic properties of FP network, we construct an FP network as a general clustering unit, and analyze its main properties. On the base of summarizing covering algorithm we present a general neighborhood covering algorithm and its corresponding network. According to the characteristics of the algorithm we analyze its existing problems and propose the further research orientations.
THz wave generated by difference-frequency in nonlinear crystal in normal temperature is theoretically studied. The conclifton of phase matching is deduced and an experimental setup is proposed. With this method, THz ...
详细信息
THz wave generated by difference-frequency in nonlinear crystal in normal temperature is theoretically studied. The conclifton of phase matching is deduced and an experimental setup is proposed. With this method, THz wave can be generated without rigorous condition and can work continuously for a long period of time in room temperature.
暂无评论