Search

Due to the scarcity of labeled samples, the class prototype obtained by support set samples is difficult to represent the real distribution of the whole class in metric-based few-shot image classification methods. Meanwhile, samples of the same class may also have large difference in many aspects and the large intra-class bias may make the sample features deviate from the class center. Aiming at the above problems that may seriously affect the performance, a few-shot image classification method with feature maps enhancement prototype (FMEP) is proposed. Firstly, this paper selects some similar features of the query set sample feature maps with cosine similarity and adds them to class prototypes to obtain more representative prototypes. Secondly, this paper aggregates similar features of the query set to alleviate the problem caused by large intra-class bias and makes features distribution of the same class closer. Finally, this paper compares enhanced prototypes and aggregated features which are both closer to real distribution to get better results. The proposed method is tested on four commonly used few-shot classification datasets, namely MiniImageNet, TieredImageNet, CUB-200 and CIFAR-FS. The results show that the proposed method can not only improve the performance of the baseline model, but also obtain better performance compared with the same type of methods.

It is a universally accepted strategy to modulate the soft and hard segments of polymer chains to achieve good elastomers. Different from other reports, in this work, Ti3C2 MXene was selected as the hard segment to prepare a poly(urethane) (PU) elastomer with integration of the structure and function. In this process, different effects of the hard segment and nanofiller of MXene and the transformation of the strengthening and toughening modes of PU with the hard segment of Ti3C2 were confirmed. The addition of a small amount of Ti3C2 resulted in the formation of balanced covalently crosslinked networks and increased the strength and toughness of the elastomer. Further addition of 4,4′-diaminodiphenyl disulfide (4AD) improved its toughness by enhancing its hydrogen bonding. The resulting PU exhibited exceptional elastomeric properties with a tensile strength of 22.1 MPa and toughness of 334.3 MJ m−3. The dynamic breaking and reformation of hydrogen bond motifs and the dynamic dissociation of disulfide bonds endowed the PU elastomer with an excellent damping capacity (about 65%), crack tolerance, and self-healing properties. After 6 hours of contact at 60 °C, the healing efficiency of two fracture sections approached 89.5%. Additionally, the good electrical conductivity of the hard segment Ti3C2 and the enhanced permittivity endowed the elastomer with excellent triboelectric performance. This work provides a general method to synthesise excellent PU elastomers with the integration of the structure and function by only modulating the hard segments. [ABSTRACT FROM AUTHOR]

The purpose of this study is to explore the relationship between the domestic patent application and international technology innovation in an open economy. Using the panel data of Asia-Pacific countries, this paper empirically analyzes the effect of economic openness level on international technological innovation and the moderating mechanism of the domestic patent application on them. The results show that the promotion effect of economic openness level on international technological innovation is different in countries with different income levels. With the increase of Foreign Direct Investment (FDI), the domestic patent application displays a promotion effect on Patent Cooperation Treaty (PCT) application. Due to the improvement of intellectual property protection in middle-high-income countries, more and more foreign investors are attracted to invest in those countries and then learn from the new technologies brought by foreign capital. The technology spillover effect of FDI is more obvious in middle-high-income countries. With the increase of Outward Foreign Direct Investment (OFDI), the domestic patent application can promote PCT application in high-income and middle-high-income countries. But there is no evidence that domestic patent application positively or negatively moderates the relationship between OFDI and international technological innovation in middle-low-income countries. [ABSTRACT FROM AUTHOR]

Stem strength plays a crucial role in the growth and development of plants, as well as in their flowering and fruiting. It not only impacts the lodging resistance of crops, but also influences the ornamental value of ornamental plants. Stem development is closely linked to stem strength; however, the roles of the SPL transcription factors in the stem development of herbaceous peony (Paeonia lactiflora Pall.) are not yet fully elucidated. In this study, we obtained and cloned the full-length sequence of PlSPL14, encoding 1085 amino acids. Quantitative real-time PCR (qRT-PCR) analysis revealed that the expression level of PlSPL14 gradually increased with the stem development of P. lactiflora and was significantly expressed in vascular bundles. Subsequently, utilizing the techniques of virus-induced gene silencing (VIGS) and heterologous overexpression in tobacco (Nicotiana tabacum L.), it was determined that PlSPL14-silenced P. lactiflora had a thinner xylem thickness, a decreased stem diameter, and weakened stem strength, while PlSPL14-overexpressing tobacco resulted in a thicker xylem thickness, an increased stem diameter, and enhanced stem strength. Further screening of the interacting proteins of PlSPL14 using a yeast two-hybrid (Y2H) assay revealed an interactive relationship between PlSPL14 and PlSLR1 protein, which acts as a negative regulator of gibberellin (GA). Additionally, the expression level of PlSLR1 gradually decreased during the stem development of P. lactiflora. The above results suggest that PlSPL14 may play a positive regulatory role in stem development and act in the xylem, making it a potential candidate gene for enhancing stem straightness in plants. [ABSTRACT FROM AUTHOR]

The diffusion and wide application of green technologies are important means to achieve climate change mitigation. Concerning the green technology absorption capacity and technology diffusion willingness of enterprises, this paper divided enterprises into four categories, constructed a green technology diffusion model and analysed the equilibrium point and stability of the model. Then, we studied the diffusion process and laws of green technology through simulation experiments. The results show that in the green technology diffusion system, the basic regeneration number is the key factor in determining the equilibrium state of green technology diffusion. The disease-free equilibrium point of the green technology diffusion system is globally asymptotically stable, and the endemic equilibrium point is locally asymptotically stable. In order to improve the speed and scale of green technology diffusion, it is necessary to create green technology diffusion sources through multiple channels, improve enterprise technology absorption capacity through multiple methods, and enhance enterprise green technology diffusion willingness at multiple levels. [ABSTRACT FROM AUTHOR]

When considering the new classification information provided by the candidate features, the features selection algorithm based on mutual information usually ignores that the addition of the candidate features will cause the change of the correlation between the selected features and class labels, which will bring additional new information; in addition, when calculating redundant information, using the form of cumulative summation may lead to the underestimation of the redundancy degree of candidate features. In view of the above problems, the definition of dynamic relevance weight is proposed to more comprehensively consider the new information components brought by candidate features. The improved definition of redundant items is proposed, and the maximum and normalization strategy are adopted to solve the problem of underestimating redundancy. On this basis, the feature selection using dynamic relevance weight (FSDRW) is proposed. Five current mainstream filter-based feature selection algorithms based on mutual information are selected for comparative experiments. Experiments on machine learning test datasets from UCI (University of California Irvine) and ASU (Arizona State University) show that the proposed algorithm works well in classification accuracy and comprehensive performance. Finally, the proposed algorithm is applied to the recognition of microseismic and blasting signals in a reservoir project in Guangxi. The selected features of the algorithm are used for microseismic signal recognition, achieving a classification accuracy of 98.86%, verifying the effectiveness of the algorithm in practical applications. [ABSTRACT FROM AUTHOR]