A video retrieval system user hopes to find relevant information when the proposed queries are ambiguous. The retrieval process based on detecting concepts remains ineffective in such a situation. Potential relationships between concepts have been shown as a valuable knowledge resource that can enhance the retrieval effectiveness, even for ambiguous queries. Recent researches in multimedia retrieval have focused on ontology modeling as a common framework to manage knowledge. Handling these ontologies has to cope with issues related to generic knowledge management and processing scalability. Considering these issues, we suggest a context-based fuzzy ontology framework for video content analysis and indexing. In this paper, we focused on the way in which we modeled our fuzzy ontology: First, we populate automatically the generated ontology by gathering various available video annotation datasets. Then, the ontology content was used to infer enhanced video semantic interpretation. Finally, considering user feedback, the content of the ontology was improved. Experimental results showed that our approach achieves the goal of scalability while at the same time allowing better video content semantic interpretation. 相似文献
Catalytic decomposition of methane (CDM) generates clean hydrogen and carbon nanomaterials. In this study, methane decomposition to hydrogen and carbon was investigated over Ni-, Co-, or Mn-doped Fe/MgO catalysts. The doping effect of different metals, varying from 3 to 10?wt%, was investigated. The catalytic performance of the obtained materials (noted 15%Fe+x%metal/MgO) revealed that the doping effect of Ni, Co, and Mn significantly improved the activity of Fe/MgO. Among the Ni-doped catalyst series, the 15%Fe+3%Ni/MgO catalyst performed better than the rest of the Ni catalysts. The 6%Co-containing catalyst remained the best in terms of activity in the Co-doped catalyst series and the 15%Fe+6%Mn/MgO solid showed better methane conversion for the Mn-doped series. Overall, 3%Ni-containing catalyst displayed the best catalytic performance among all Ni-, Co-, and Mn-doped catalysts. XRD, N2 sorption, and H2 temperature-programmed reduction (TPR), Laser–Raman spectroscopy, thermogravimetric analysis (TGA) under air, and temperature-programmed oxidation (TPO) were used for catalyst characterization. The results revealed that all the doped catalysts exhibited better metallic active site distribution than 15%Fe/MgO and proved that metal doping played a crucial role in catalytic performance. 相似文献
Bulk Co samples having a mean grain size of ~300 nm were processed by hot isostatic pressing of a high purity Co nanopowder synthesized by chimie douce. The grain interior exhibited a highly faulted nanoscale lamellar microstructure comprising an intricate mixture of face-centered cubic, hexagonal close-packed phases and nanotwins. Room temperature compression tests carried out at a strain rate of ~2 × 10?4 s?1 revealed a yield stress of ~1 GPa, a strain to rupture of ~5%. During straining it was found that the hexagonal close-packed phase content increased from 55% to 65% suggesting a deformation mechanism based on stress-assisted face-centered cubic to hexagonal close-packed phase transformation. In addition, an apparent activation volume of ~3b3 was computed which indicates that the deformation mechanism was controlled by dislocation nucleation from the numerous boundaries. Nonetheless, in such an intricate microstructure, the overall mechanical properties are discussed in term of a complex interplay between lattice dislocation plasticity, transformation-induced plasticity and possibly twin-induced plasticity. 相似文献
Nanocrystalline (Ti0.8Zr0.2)C powder consisting in grains of about 200 nm in diameter obtained by mechanical alloying was sintered by a spark plasma sintering (SPS) process without the addition of any binder phase. The microstructure, Vickers micro hardness and density in relation to the sintering time and temperature are carefully described. The most suitable sintering condition under pressure of 50 MPa is 1650 °C for 5 min. In this sintering condition, the hardness can reach 2760 Hv and the relative density can reach 98%. 相似文献
Usually, a large number of reference signatures are required for building the writing style model from offline handwritten signatures (OHSs). Moreover, the existing writer identification systems from OHSs are generally closed systems that require a retraining process when a new writer is added. This paper proposes an open writer identification system from OHSs, based on a new scheme of the one-class symbolic data analysis (OC-SDA) classifier, using few reference signatures. For generating more data, intra-class feature-dissimilarities, generated from curvelet transform, are introduced for building the symbolic representation model (SRM) associated with each writer. Feature-dissimilarities allow capturing more efficiently the intra-personnel variability produced naturally by a writer and, thus, increase the inter-personnel variability. Instead of using the mean and the standard deviation for building the OC-SDA model, intra-class feature-dissimilarities generated for each writer are modeled through a new weighted membership function, inspired from the real probability distribution of training intra-class feature-dissimilarities. The comparative analysis against the state-of-the-art works shows that the proposed OC-SDA classifier outperforms the existing classifiers on three public signature datasets GPDS-300, CEDAR-55 and MCYT-75, using only five reference signatures, achieving 98.31%, 98.06% and 99.89%, respectively, even when a combination of multiple classifiers is performed or even using learned features. Moreover, the evaluation of the proposed writer identification system in front of skilled forgeries shows its ability to detect also possible forged signatures in addition to the genuine ones.
Neural Computing and Applications - Automatic facial expression recognition (FER) is one of the most challenging tasks in computer vision. FER admits a wide range of applications in... 相似文献
A robust algorithm based on Twin Support Vector Regression and discrete wavelet transform applied to millimetric wave (mmWave) channel prediction is proposed in this work. The 60 GHz band is appropriate for small-scale high-speed data transmission applications in future 5G indoor network solutions. The experimentation takes place in an enclosed complex conference room setting with furniture and computer equipment. The proposed algorithm is applied to mmWave multipath channel with higher order modulation scheme with receiver sensitivity thresholds being ? 80 dBm, ? 90 dBm, ? 100 dBm and ? 110 dBm corresponding to 41, 89, 195 and 250 paths, respectively. The Channel Impulse Response of 60 GHz multipath wireless system is generated by the “Wireless InSite” ray tracer by Remcom. Compared to other traditional algorithms, numeric experiments confirm the effectiveness of the proposed solution under several multipath configurations.