This paper describes a new method for active learning in content-based image retrieval. The proposed method firstly uses support vector machine (SVM) classifiers to learn an initial query concept. Then the proposed ac...This paper describes a new method for active learning in content-based image retrieval. The proposed method firstly uses support vector machine (SVM) classifiers to learn an initial query concept. Then the proposed active learning scheme employs similarity measure to check the current version space and selects images with maximum expected information gain to solicit user's label. Finally, the learned query is refined based on the user's further feedback. With the combination of SVM classifier and similarity measure, the proposed method can alleviate model bias existing in each of them. Our experiments on several query concepts show that the proposed method can learn the user's query concept quickly and effectively only with several iterations.展开更多
We numerically study the dynamics of meandering spiral waves in the excitable system subjected to a feedback signal coming from two measuring points located on a straight line together with the initial spiral core. Th...We numerically study the dynamics of meandering spiral waves in the excitable system subjected to a feedback signal coming from two measuring points located on a straight line together with the initial spiral core. The core location and size radius of the final attractors are computed, and they change with the position of the moving measuring point in a unique way. By the Fourier Spectral analysis, we find the frequency-locked behaviors different from the driving scheme of the external periodic force. It when the moving measuring point approaches closely is also found that the meandering spiral wave can be eliminated the boundary and its feedback gain is large enough. This offers an effective and convenient method for eliminating meandering spiral waves.展开更多
文摘This paper describes a new method for active learning in content-based image retrieval. The proposed method firstly uses support vector machine (SVM) classifiers to learn an initial query concept. Then the proposed active learning scheme employs similarity measure to check the current version space and selects images with maximum expected information gain to solicit user's label. Finally, the learned query is refined based on the user's further feedback. With the combination of SVM classifier and similarity measure, the proposed method can alleviate model bias existing in each of them. Our experiments on several query concepts show that the proposed method can learn the user's query concept quickly and effectively only with several iterations.
基金Supported by the National Natural Science Foundation of China under Grant No. 11005030Science Foundations of Hebei Education Department under Grant No. 2009135+1 种基金Science Foundations of Inner Mongolia Education Department under Grant No. NJ09178Science Foundation of Hebei Normal University
文摘We numerically study the dynamics of meandering spiral waves in the excitable system subjected to a feedback signal coming from two measuring points located on a straight line together with the initial spiral core. The core location and size radius of the final attractors are computed, and they change with the position of the moving measuring point in a unique way. By the Fourier Spectral analysis, we find the frequency-locked behaviors different from the driving scheme of the external periodic force. It when the moving measuring point approaches closely is also found that the meandering spiral wave can be eliminated the boundary and its feedback gain is large enough. This offers an effective and convenient method for eliminating meandering spiral waves.
