Search

In computer vision and pattern recognition applications, there are usually a vast number of unlabelled data whereas the labelled data are very limited. Active learning is a kind of method that selects the most representative or informative examples for labelling and training; thus, the best prediction accuracy can be achieved. A novel active learning algorithm is proposed here based on one-versus-one strategy support vector machine (SVM) to solve multi-class image classification. A new uncertainty measure is proposed based on some binary SVM classifiers and some of the most uncertain examples are selected from SVM output. To ensure that the selected examples are diverse from each other, Gaussian kernel is adopted to measure the similarity between any two examples. From the previous selected examples, a batch of diverse and uncertain examples are selected by the dynamic programming method for labelling. The experimental results on two datasets demonstrate the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Potassium hydride (KH) was directly added to a Mg(NH2)2-2LiH system to improve the hydrogen storage properties; the corresponding mechanisms were elucidated. The Mg(NH2)2-2LiH-0.08KH composite displays optimized hydrogen-storage properties, reversibly storing approximately 5.2 wt% hydrogen through a two-stage reaction and a dehydrogenation onset at 70 °C. The 0.08KH-added sample fully dehydrogenated at 130 °C begins to absorb hydrogen at 50 °C, and takes up approximately 5.1 wt% of hydrogen at 140 °C. Adding KH significantly enhances the de-/hydrogenation kinetic properties; however, an overly rapid hydrogenation rate enlarges the particle size and raises the dehydrogenation temperature. A cycling evaluation reveals that the KH-added Mg(NH2)2-2LiH system possesses good reversible hydrogen storage abilities, although the operational temperatures for de-/hydrogenation increase during cycling. Detailed mechanistic investigations indicate that adding KH catalytically decreases the activation energy of the first dehydrogenation step and reduces the enthalpy of desorption during the second dehydrogenation step as a reactant, significantly improving the hydrogen storage properties of Mg(NH2)2-2LiH. [ABSTRACT FROM AUTHOR]

The effects of ionic strength on the adsorption behaviors of the monovalent cation NpO2+, the divalent cations UO22+, Zn2+, Sr2+, the trivalent cations Yb3+, Am3+ onto Al2O3, SiO2 and Fe2O3 from aqueous solutions were investigated. It was found that depending on the adsorption system and solution condition, the ionic strength effects are promotive, reductive and insensitive to increasing ionic strength. Based on the comparison among the various effects of ionic strength on the various adsorption systems and the theoretical considerations, it may be concluded that the factors of ionic strength effect on cation adsorption are very complicated, that there are various effects on a certain adsorption system in the different pH ranges and different concentration ranges of background electrolyte, that the speciation of metals in aqueous solutions plays an important role in the cation adsorption, that the coulombic interaction and the chemical reaction could be directed towards the various ionic strength effects and that the satisfactory interpretation of various effects of ionic strength is very difficult to access and that ionic strength effect on cation adsorption is an open question and worthy of further study, because of the great practical importance and theoretical interest. [Copyright &y& Elsevier]