In some image classification tasks, similarities among different categories are different and the samples are usually misclassified as highly similar categories. To distinguish highly similar categories, more specific...In some image classification tasks, similarities among different categories are different and the samples are usually misclassified as highly similar categories. To distinguish highly similar categories, more specific features are required so that the classifier can improve the classification performance. In this paper, we propose a novel two-level hierarchical feature learning framework based on the deep convolutional neural network(CNN), which is simple and effective. First, the deep feature extractors of different levels are trained using the transfer learning method that fine-tunes the pre-trained deep CNN model toward the new target dataset. Second, the general feature extracted from all the categories and the specific feature extracted from highly similar categories are fused into a feature vector. Then the final feature representation is fed into a linear classifier. Finally, experiments using the Caltech-256, Oxford Flower-102, and Tasmania Coral Point Count(CPC) datasets demonstrate that the expression ability of the deep features resulting from two-level hierarchical feature learning is powerful. Our proposed method effectively increases the classification accuracy in comparison with flat multiple classification methods.展开更多
In this study, we compared two types of EEG modalities, sensory-motor rhythms(SMR) and movement related cortical potentials(MRCP), on four healthy subjects performing ballistic or repetitive movement imagination. The ...In this study, we compared two types of EEG modalities, sensory-motor rhythms(SMR) and movement related cortical potentials(MRCP), on four healthy subjects performing ballistic or repetitive movement imagination. The EEG waveform morphology across subjects was similar for MRCPs, whereas there was not a clear pattern for SMRs. The rank-sum test showed a significant difference between the amplitude of baseline and that of the MRCP as early as 2 s prior to imagery onset, for both types of motor imageries, indicating strong discriminative power of MRCPs for predicting movement onset. For SMR, this type of discriminative power was relatively weak and highly subject-specific. On the other hand, the SMR landscape under the two movement imagery types was distinctive, holding a potential for discriminating the two movement imagery types. These preliminary results presented different characteristics of SMR and MRCP under different motor imageries, providing valuable information regarding the design and implementation of motor imagery based on BCI system.展开更多
基金Project supported by the National Natural Science Foundation of China(No.61379074)the Zhejiang Provincial Natural Science Foundation of China(Nos.LZ12F02003 and LY15F020035)
文摘In some image classification tasks, similarities among different categories are different and the samples are usually misclassified as highly similar categories. To distinguish highly similar categories, more specific features are required so that the classifier can improve the classification performance. In this paper, we propose a novel two-level hierarchical feature learning framework based on the deep convolutional neural network(CNN), which is simple and effective. First, the deep feature extractors of different levels are trained using the transfer learning method that fine-tunes the pre-trained deep CNN model toward the new target dataset. Second, the general feature extracted from all the categories and the specific feature extracted from highly similar categories are fused into a feature vector. Then the final feature representation is fed into a linear classifier. Finally, experiments using the Caltech-256, Oxford Flower-102, and Tasmania Coral Point Count(CPC) datasets demonstrate that the expression ability of the deep features resulting from two-level hierarchical feature learning is powerful. Our proposed method effectively increases the classification accuracy in comparison with flat multiple classification methods.
文摘In this study, we compared two types of EEG modalities, sensory-motor rhythms(SMR) and movement related cortical potentials(MRCP), on four healthy subjects performing ballistic or repetitive movement imagination. The EEG waveform morphology across subjects was similar for MRCPs, whereas there was not a clear pattern for SMRs. The rank-sum test showed a significant difference between the amplitude of baseline and that of the MRCP as early as 2 s prior to imagery onset, for both types of motor imageries, indicating strong discriminative power of MRCPs for predicting movement onset. For SMR, this type of discriminative power was relatively weak and highly subject-specific. On the other hand, the SMR landscape under the two movement imagery types was distinctive, holding a potential for discriminating the two movement imagery types. These preliminary results presented different characteristics of SMR and MRCP under different motor imageries, providing valuable information regarding the design and implementation of motor imagery based on BCI system.
