基于最小方差的声线跟踪自适应分层算法

针对高精度水声定位系统中定位精度与计算量相互矛盾的问题,提出了基于最小方差的声线跟踪自适应分层算法。该算法根据声速在有限范围内规律变化的特点,对声速剖面全局搜索,通过各段声速散点到其拟合直线的总方差表示拟切分段与原始声速剖面的差别,利用最大差别找出最佳切分点,保留原始声速剖面特征,实现对声速剖面的简化。实验结果表明,该算法在保证定位精度的同时能够提高计算效率,具有良好的工程应用价值。     

基于人脸卡口数据的行人共现关系图谱构建

随着人脸识别技术的不断进步以及人脸卡口的大范围且密集的部署, 本文针对团伙犯罪案件侦察这一应用场景, 对人脸卡口数据进行深入挖掘, 探究其中行人间的共现关系, 获取所关注的嫌疑人的现实社交网络, 锁定团伙其余人员. 经过实验比对和论证, 本文使用Chinese Whispers聚类算法对行人节点进行识别, 通过Faiss加速邻接边的构建, 加速图的初始化步骤, 解决其聚类效率低下的问题. 在此基础上, 使用共现频次和Apriori算法中的置信度挖掘行人间的共现关系, 构建行人共现关系图谱.     

High-temperature ferromagnetism of Ni-doped PbPdO2 nanograin films synthesized by sol-gel spin-coating method

The single-phase Ni-doped PbPdO₂ films with a body-centered orthorhombic structure were synthesized by the sol-gel spin-coating method and an oxidation treatment. The films with a thickness of about 440 nm were found to have a nanograin structure and large amount of Pb vacancies. The valence states of Ni and Pd ions within the Ni-doped PbPdO₂ films were very close to 2+, while the Pb ions exhibited a mixed valence between 2+ and 4+. The existence of Pb vacancies and the low electronegativity of Pb²⁺ ion resulted in the increase of the valence of Pb ions. The analysis of the magnetic properties indicated that the magnetisms of the single-phase Ni-doped PbPdO₂ nanograin films were all composed of the ferromagnetism and the paramagnetism. The ferromagnetism enhanced with the calcination temperature increasing and could be retained up to 380 K. A carrier-mediated mechanism bridged to the bound magnetic polaron model based on the Pb vacancies, the doped Ni ions and the Pb ions with a valence higher than 2+ were used to explain the magnetic origin of these Ni-doped PbPdO₂ nanograin films.     

Content-based filtering for recommendation systems using multiattribute networks

Content-based filtering (CBF), one of the most successful recommendation techniques, is based on correlations between contents. CBF uses item information, represented as attributes, to calculate the similarities between items. In this study, we propose a novel CBF method that uses a multiattribute network to effectively reflect several attributes when calculating correlations to recommend items to users. In the network analysis, we measure the similarities between directly and indirectly linked items. Moreover, our proposed method employs centrality and clustering techniques to consider the mutual relationships among items, as well as determine the structural patterns of these interactions. This mechanism ensures that a variety of items are recommended to the user, which improves the performance. We compared the proposed approach with existing approaches using MovieLens data, and found that our approach outperformed existing methods in terms of accuracy and robustness. Our proposed method can address the sparsity problem and over-specialization problem that frequently affect recommender systems. Furthermore, the proposed method depends only on ratings data obtained from a user's own past information, and so it is not affected by the cold start problem.     

Critical parameters in cost-effective alkaline extraction for high protein yield from leaves

Leaves are potential resources for feed or food, but their applications are limited due to a high proportion of insoluble protein and inefficient processing. To overcome these problems, parameters of alkaline extraction were evaluated using green tea residue (GTR). Protein extraction could be maximized to 95% of total protein, and, after precipitation by pH adjustment to 3.5, 85% of extracted protein was recovered with a purity of 52%. Temperature, NaOH amount, and extraction time are the protein yield determining parameters, while pH and volume of extraction liquid are critical parameters for production cost. The cost of energy and chemicals for producing 1 t GTR proteins is minimized to 102€, and its nutritional value is comparable to soybean protein. Furthermore, this technology was successfully applied to other sources of biomass and has potential to be used as a part of an integrated bio-refinery process.     

Influence of Si-Doping on the Performance of InGaN/GaN Multiple Quantum Well Solar Cells

Semiconductors - The performance of InGaN/GaN multiple quantum well (MQW) solar cells with five different Si-doping concentrations, namely 0, 4 × 1017 cm–3, 1 × 1018 cm–3, 3...     

Aflatoxin B1 degradation by culture and lysate of a Pontibacter specie

The presence of aflatoxin B₁ (AFB₁) along the food chain poses a significant threat, thus propelling the need for an effective approach to control it. This study was therefore, aimed at investigating AFB₁ degradation of liquid cultures and lysates of an isolated Pontibacter sp. (VGF1). Liquid cultures, lysed bacterial cells in the absence (uninhibited lysates) and presence of protease inhibitors (protease inhibited lysates) were respectively incubated with AFB₁ for 3, 6, 12, 24 and 48 h. AFB₁ degradation was monitored during this period on high performance liquid chromatography (HPLC) and results obtained revealed that after 6 h of incubation, the protease inhibited (PI) lysates yielded a 65% AFB₁ degradation, whereas after 12 h, no residual AFB₁ was detected. Conversely, after 48 h of incubation, a significantly (p≤0.05) lower AFB₁ degradation of 50 and 36% by the liquid culture and uninhibited lysate, respectively, were noted. It was further confirmed that the degradation mechanism was enzymatic. Data from cytotoxicity studies against human lymphocytes further demonstrated that extracts of biotransformed AFB₁ were less toxic when compared to that of AFB₁. Findings from this study have demonstrated an alternative approach for the decontamination and biocontrol of AFB₁ in various agricultural commodities.