Sentinel-2卫星的多光谱轻量级船舶目标检测算法

近年来深度卷积神经网络在可见光船舶检测方面取得了显著的进展，然而，大多数相关研究是通过改进大型的网络结构来提高检测性能，因此加大了对更高计算机性能的需求。此外，可见光图像难以在云、雾、海杂波、黑夜等复杂场景检测到船舶。针对以上问题，提出了一种融合红（red, R）、绿（green, G）、蓝（blue, B）和近红外（NIR）4个波段光谱信息的由粗到精细的轻量型船舶检测算法。与现有的方法中根据光谱特性利用水体检测算法提取水体区域不同之处是该算法是利用改进的水体检测算法来提取船舶候选区域。为获取更准确的候选区域，对船舶、厚云、薄云、平静海面、杂波海面5种场景中4个波段的像素值进行了统计分析，选取近红外大于阈值作为辅助判断，并以其中心点获取候选区域32×32大小的切片，并对切片进行非极大值抑制，由此获得了船舶粗检测结果。随后构建了轻量级LSGFNet网络对船舶候选区域切片进行精细识别。构建的网络融合了1×1卷积提取的波谱特征与3×3的提取几何特征，为防止光谱特征与几何特征的信息在融合时“信息不流通”，在LSGFNet网络中引入了ShuffleNet中的通道打乱机制，并减小了模型结构，与典型的轻量级网络相比具有更好的效果且模型较小。最后，利用Sentinel-2卫星多光谱10 m分辨率数据构建了512×512大小的1 120组数据进行粗检测，以及32×32大小的6 014组数据进行精细网络训练，其中候选区域粗提取的查全率为98.99%，精细识别网络精确度为96.04%，不同场景下的平均精确度为92.98%。实验表明该算法在抑制云层、海浪杂波等干扰的复杂背景下具有较高的检测效率，且训练时间短、计算机性能需求低。     

Highly sensitive perylene tetra-(alkoxycarbonyl) based colorimetric and ratiometric probes for fluorine detection

Fluoride anion (F^?) affects environmental, biological, and chemical processes significantly. Therefore, its detection has received increasing attention, and sensitive, effective, and convenient probes for F^? detection need to be developed urgently. In this work, two perylene tetra-(alkoxycarbonyl) (PTAC) based colorimetric and ratiometric probes, P1 and P2, were developed for the detection of F^?. The interactions between F^? and these two probes were investigated by absorption, electrochemistry, ¹H NMR, and density functional methods. Both the two probes were complexed with F^? with a ratio of 1:1. The detection limits of P1 and P2 were 0.22 μM and 0.87 μM, respectively. It was worth noting that the absorption peak of P1 showed a 190 nm red shift when sensing F^?, and P1 is the largest red shift value reported in F^? probes based on PTAC derivatives. This phenomenon was resulted from the unique configuration and deprotonation of P1 that can promote the intramolecular charge transfer (ICT). This strategy provides an example for the development of other ion probes based on D-A type ICT mechanism.     

Mg,N掺杂β-Ga2O3光电性质的第一性原理计算

通过基于密度泛函理论的第一性原理计算,研究了Mg单掺杂、N单掺杂和不同浓度的Mg-N共掺杂β-Ga₂O₃的结构性质、电子性质和光学性质,以期获得性能比较优异的p型β-Ga₂O₃材料。建立了五种模型:Mg单掺杂、N单掺杂、1个Mg-N共掺杂、2个Mg-N共掺杂和3个Mg-N共掺杂β-Ga₂O₃。经过计算,3个Mg-N共掺杂β-Ga₂O₃体系的结构最稳定。此外,在5种模型中,3个Mg-N共掺杂β-Ga₂O₃体系的禁带宽度是最小的,并且N 2p和Mg 3s贡献的占据态抑制了氧空位的形成,从而增加了空穴浓度。因此,3个Mg-N共掺杂β-Ga₂O₃体系表现出优异的p型性质。3个Mg-N共掺杂体系的吸收峰出现明显红移,在太阳盲区的光吸收系数较大,这归因于导带Ga 4s、Ga 4p、Mg 3s向价带O 2p、N 2p的带间电子跃迁。本工作将为p型β-Ga₂O₃日盲光电材料的研究和应用提供理论指导。     

Application of advanced nuclear analytical techniques for the electrocatalyst's characterization: Paving the path for mechanistic investigations

As the application of electrocatalyst continues to expand, envisaging the hidden mechanisms occurring at various length scale affecting the catalytic efficiency became important. To enhance the stability of electrocatalyst and reduce the cost, it is of paramount importance to reveal the active site's dynamics (using in situ techniques for getting the real-time information) which directly affect the reactions such as oxygen evolution reaction, hydrogen evolution reaction, and so on. Since such reactions are crucial for many engineering and scientific applications, in situ characterization techniques are required, which could capture such reactions happening at a different length and time scale. This article analyzes the recent progress made in the field of electrocatalyst's characterization using in situ neutron techniques. The article also paves the future path and has delineated the future challenges involved in multiscale correlative techniques (e.g., neutron techniques in the combination of synchrotron or microscopic techniques) used for getting the multiscale (atomic to micrometer range) mechanistic information about the electrocatalyst's working and degradation.     

Modifying the physicochemical properties,solubility and foaming capacity of milk proteins by ultrasound-assisted alkaline pH-shifting treatment

This study investigated the effects of different treatment of alkaline pH-shifting on milk protein concentrate (MPC), micellar casein concentrate (MCC) and whey protein isolate (WPI) assisted by the same ultrasound conditions, including changes in the physicochemical properties, solubility and foaming capacity. The solubility of milk proteins had a significant increase with gradual enhancement of ultrasound-assisted alkaline pH-shifting (p < 0.05), especially for MCC up to 99.50 %. Also, treatment made a significant decline in the particle size of MPC and MCC, as well as the turbidity of the proteins (p < 0.05). The foaming capacity of MPC, MCC, and WPI was all improved, especially at pH 11, and at this pH, the milk protein also showed the highest surface hydrophobicity. The best foaming capacity at pH 11 was the result of the combined effect of particle size, potential, protein conformation, solubility, and surface hydrophobicity. In conclusion, ultrasound-assisted pH-shifting treatment was found to be effective in improving the physicochemical properties and solubility and foaming capacity of milk proteins, especially MCC, with promising application prospect in food industry.     

Regulating Phase Junction and Oxygen Vacancies of TiO2 Nanoarrays for Boosted Photoelectrochemical Water Oxidation

In this study, we have provided a facile solution to synthesize well-aligned titanium dioxide nanorods by using hydrothermal reaction. By calcining the materials under different atmospheres and temperatures, a batch of titanium dioxides with excellent oxygen evolution reaction(OER) catalytic efficiency were obtained. This new structured TiO₂ photoanode material yields a high photocurrent density of 5.69 mA/cm² at 1.23 V vs. reversible hydrogen electrode(RHE) under simulated solar light(100 mW/cm²). Surface photovoltage techniques and other measurements were carried out to confirm that the enhanced photoelectrochemical performances were attributed to the synergistic effect of the phase junction and a certain content of surface states, which accelerate the separation and transmission of the photogenerated charges. This material with phase junction and surface states promises a potential application in the field of photoelectric catalysis under solar light.     

A Stable Porphyrin Functionalized Graphite Electrode Used at the Oxygen Evolution Reaction Potential

Many researches have been devoted to rechargeable power generators that can store (but also release) energy. This availability is ensured through (e. g.) the oxygen evolution reaction (OER). However, (i) large values of the overpotentials and (ii) a progressive detriment of the anode (graphite) electrode limit the ultimate device. In view of enhancing the electrode performances, graphite was protected by following different strategies, which oblige to follow precise preparation protocols. Here, we prove that a thin layer of free-base porphyrin molecules is able to protect the underneath graphite electrode from detriment even if many (about 100) electrochemical cycles are performed.     

紫色球杆菌视紫红质光谱特性的机器学习研究

近年来，机器学习等人工智能技术被应用于蛋白质工程，其在蛋白质结构、功能预测、催化活性等研究中具有独特优势。在未知蛋白质结构的情况下，将蛋白质序列和功能特性与机器学习相结合，基于序列-活性关系（innovative sequence-activity relationship，ISAR）算法，将蛋白质氨基酸序列数字化，用快速傅里叶变换（fast four transform，FFT）进行预处理，再进行偏最小二乘回归建模，可在数据集较少情况下拟合得到最佳模型。通过机器学习对紫色球杆菌视紫红质（gloeobacter violaceus rhodopsin，GR）的突变体蛋白质氨基酸序列与光谱最大吸收波长进行建模，获得了最佳模型。用最佳索引LEVM760106建模得到的确定系数R² 为0.944，均方误差E为11.64。用小波变换进行的预处理，其R² 虽也约为0.944，但E大于11.64，不及FFT进行的预处理。方法较好地解决了蛋白质序列与功能特性之间的数学建模问题，在蛋白质工程中可为预测更优的突变体提供支持。