Adaptive reversible data hiding for JPEG images with multiple two-dimensional histograms

Joint photographic experts group (JPEG) can provide good quality with small file size but also eliminate extensively the redundancies of images. Therefore, hiding data into JPEG images in terms of maintaining high visual quality at small file sizes has been a great challenge for researchers. In this paper, an adaptive reversible data hiding method for JPEG images containing multiple two-dimensional (2D) histograms is proposed. Adaptability is mainly reflected in three aspects. The first one is to preferentially select sharper histograms for data embedding after $K$ histograms are established by constructing the $k$th ($k\in \{1,2,\dots ,K\}$) histogram using the $k$th non-zero alternating current (AC) coefficient of all the quantized discrete cosine transform blocks. On the other hand, to fully exploit the strong correlation between coefficients of one histogram, the smoothness of each coefficient is estimated by a block smoothness estimator so that a sharply-distributed 2D-histogram is constructed by combining two coefficients with similar smoothness into a pair. The pair corresponding to low complexity is selected priorly for data embedding, leading to high embedding performance while maintaining low file size. Besides, we design multiple embedding strategies to adaptively select the embedding strategy for each 2D histogram. Experimental results demonstrate that the proposed method can achieve higher rate–distortion performance which maintaining lower file storage space, compared with previous studies.     

Semi-supervised Learning with Graph Convolutional Networks Based on Hypergraph

Neural Processing Letters - Graph convolutional networks (GCNs), which rely on graph structures to aggregate information of neighbors to output robust node embeddings, have been becoming a popular...     

Revealing the role of Ni2+ ions in inducing the synthesis of porous carbon balls: A novel substrate to enhance the Pt catalytic activity towards methanol-oxidation

Carbon-based materials have been often employed as electrocatalytic substrates because of their large surface area/highly porous structure. Similar to carbon substrates, the non-carbon related materials such as transition metals also play an important role in improving catalytic performance. However, the simultaneous synthesis and metallic functionalization of carbon substrates is a highly challenging issue. Herein, a hydrothermal method has been used for the preparation of Ni-functionalized porous carbon balls. The significant role of Ni²⁺ ions in the synthesis of porous carbon balls has been confirmed. The results of transmission electron microscopy indicate that, the as-prepared porous carbon balls were suitable for the dispersion of Pt nanoparticles with small particle size (less than 4 nm). In addition to providing the OH_ads species, the Ni can also modify the surface electronic structure of Pt. Electrochemical measurements results reveal that, under the strong interactions between Ni and Pt, the as-prepared porous carbon balls supported Pt nanoparticles (Pt/Ni-CB) catalyst possesses excellent electrocatalytic activity, stability and CO anti-poisoning capability towards methanol electrooxidation reaction (MOR). This work opens a novel idea for the construction of the metal functionalization of carbon substrates and their subsequent applications in other electrocatalytic reactions.     

Adaptive Graph Learning for Semi-supervised Self-paced Classification

Neural Processing Letters - Semi-supervised learning techniques have been attracting increasing interests in many machine learning fields for its effectiveness in using labeled and unlabeled...     

低共熔溶剂用于CO2捕集的研究进展

低共熔溶剂(DESs)具有原料廉价易得、化学稳定性好、可设计、合成工艺简单、可循环使用和绿色环保等优点,在CO₂捕集领域受到广泛关注。重点综述了近年来DESs用于CO₂捕集的研究进展,总结了DESs捕集CO₂的能力,分析了DESs捕集CO₂的影响因素和DESs的循环使用性能,归纳了DESs捕集CO₂的机理(包括物理吸收、化学吸收和物理化学协同吸收),并总结了CO₂在DESs中的溶解度计算模型。分析发现,DESs捕集CO₂的影响因素中,温度、压力和水含量(质量分数)均对CO₂的捕集有影响,且DESs的结构是重要的影响因素;大部分DESs可循环使用;CO₂在DESs中溶解度计算模型的建立有效推动了DESs捕集CO₂的进一步发展。最后,指出了DESs捕集分离CO₂所面临的主要问题并对进一步的研究工作进行了分析讨论。     

Synthesis of transparent YIG ceramics by pressureless sintering

Transparent YIG (Y₃Fe₅O₁₂) ceramics are successfully synthesized by reactive sintering at normal pressure using γ-Fe₂O₃ and Y₂O₃ as starting materials. The grain size of the sintered YIG ceramics is ca. 10–15 µm. Residual pores are not observed on the surface of sample, but numerous residual pores are observed by infrared transmission microscopy. In-line transmittance of a commercially available high-quality YIG single crystal (thickness 1 mm) fabricated by the floating zone method is 75 % in the near to mid-infrared region, whereas the sample produced in this study shows an in-line transmittance of 71 % in the wavelength range above 1.5 µm.     

改进的多层感知机在客户流失预测中的应用

针对在传统的客户流失预测数据预处理中，使用one-hot编码处理离散属性导致数据维度增加及数据过于稀疏的问题，提出了两种基于多层感知机的改进后的客户流失预测模型。其主要思想是分别使用堆叠自编码器和实体嵌入两种方法对多层感知机进行改进，通过将离散属性的高维编码数据向低维空间映射，有效地减少了one-hot编码产生的稀疏数据，增加了离散属性值之间的关联度。在对两份公开的数据集进行交叉验证后的实验结果表明，改进后的模型既有效地提高了预测的准确度，又维持了传统多层感知机模型在并行化计算方面的优势。     

Toward nanograined gadolinia doped ceria via two-step sintering at ultralow temperature and its electrical conductivity

Highly dense (>98%) and nanograined (∼60 nm) gadolinia doped ceria are obtained from ultrafine powders by adopting two-step sintering (TSS) procedure at an ultralow temperature of 750 °C with a dwell time of 20 h, which is the lowest sintering temperature for ceria family without sintering aids up to now. Electrochemical impedance spectroscopy investigations suggest that the electrical conductivities of densified electrolytes are closely related to sintering temperature and grain size, and GDC900-750 exhibits the highest total electrical conductivity of 3.640 S m⁻¹ at 700 °C in air. Fitting calculation indicates partial grain-size dependence of oxygen vacancy association enthalpy and grain-size independence of oxygen ion migration enthalpy. Grain boundary maturity influences on grain boundary conductivity to some extent, and younger grain boundary endues the densified electrolytes with higher grain boundary conductivity.     

ZnSe/CoSe/NCDPH composites as anode materials for lithium ion batteries with high capacity and long cycle

In this paper, dopamine hydrochloride (DPH) is introduced to synthesize ZIF-8@ZIF-67@DPH in the preparation of ZIF-8@ZIF-67. ZnSe/CoSe/NC_DPH (N-doped carbon) composites are calcined in a high-temperature inert atmosphere with ZIF-8@ZIF-67@DPH as the precursor, selenium powder as the selenium source. ZnSe/CoSe/NC_DPH has high discharge specific capacity, good cycle stability and outstanding rate performance. The first discharge capacity of ZnSe/CoSe/NC_DPH is 1616.6 mAh g⁻¹ at the current density of 0.1 A g⁻¹, and the reversible capacity remains at 1214.2 mAh g⁻¹ after 100 cycles, the reversible capacity is 416.7 mAh g⁻¹ after 1000 cycles at 1 A g⁻¹. Therefore, ZnSe/CoSe/NC_DPH composites provide a new step for the research and synthesis of new stable, high-capacity, and safe high-performance lithium ion batteries. The bimetallic selenide composites not only have bimetallic active sites, but also can form synergistic effect between different metal phases, which can effectively reduce the capacity attenuation caused by volume expansion and reactive stress enrichment during lithium storage of metal oxide anode materials. Meanwhile, N-doped carbon can improve the conductivity and provide more active sites to store lithium, thus improving its lithium storage capacity.