基于快速神经网络架构搜索的鲁棒图像水印网络算法

为解决深度学习在图像水印算法中计算量大且模型冗余的问题，提高图像水印算法在抵抗噪声、旋转和剪裁等攻击时的鲁棒性，提出基于快速神经网络架构搜索（neural architecture search，NAS）的鲁棒图像水印网络算法。通过多项式分布学习快速神经网络架构搜索算法，在预设的搜索空间中搜索最优网络结构，进行图像水印的高效嵌入与鲁棒提取。首先，将子网络中线性连接的全卷积层设置为独立的神经单元结构，并参数化表示结构单元内节点的连接，预先设定结构单元内每个神经元操作的搜索空间；其次，在完成一个批次的数据集训练后，依据神经元操作中的被采样次数和平均损失函数值动态更新概率；最后，重新训练搜索完成的网络。水印网络模型的参数量较原始网络模型缩减了92%以上，大大缩短了模型训练时间。由于搜索得到的网络结构更为紧凑，本文算法具有较高的时间性能和较好的实验效果，在隐藏图像时，对空域信息的依赖比原始网络更少。对改进前后的2个网络进行了大量鲁棒性实验，对比发现，本文算法在CIFAR-10数据集上对抵抗椒盐噪声和旋转、移除像素行（列）等攻击优势显著；在ImageNet数据集上对抵抗椒盐高斯噪声、旋转、中值滤波、高斯滤波、JPEG压缩、裁剪等攻击优势显著，特别是对随机移除行（列）和椒盐噪声有较强的鲁棒性。     

对氯苯甲酸构筑的铜(Ⅱ)配合物的合成、HSA结合及细胞毒性

本文合成了配合物[Cu(pcba)₂·(phen)(H₂O)] (pcba =对氯苯甲酸,phen = 1,10-邻菲罗啉),该配合物属于三斜晶系,P1空间群,晶胞参数为a=0.790 98(2) nm,b=1.072 40(4) nm,c=1.487 19(6) nm,α=100.613(3)°,β=95.239(3)°,γ=108.334(3)°,Z=2,D_c=1.638 g·cm^-3,F(000)=582,最终结构残差因子R₁=0.035 9,wR₂=0.089 1。采用紫外及荧光研究了配合物和人血清蛋白(HSA)的相互作用方式。结果表明,配合物静态猝灭HSA荧光,可求得配合物与HSA的猝灭常数K_sv=2.35×10⁵ L·mol^-1,猝灭速率常数K_q=2.35×10¹³ L·mol^-1·s^-1,结合常数为K_a=2.14×10¹³ L·mol^-1,结合位点n=2.37。同时,研究了配合物对胃癌细胞A549、宫颈癌细胞Hela和肝癌细胞HepG2的抗增殖能力。     

汽配件颜色喷涂顺序问题的TSP转化与建模

汽配件颜色喷涂顺序问题通常以生产线上相邻汽配件颜色切换次数少为最优目标,以进一步降低生产成本.该类问题具有所有汽配件都必须喷涂一次且只喷涂一次的特点,为此提出了TSP转化与建模的方法.将待喷涂汽配件定义为TSP顶点,任意两个待喷涂汽配件的颜色切换定义为顶点的距离,仿照TSP问题构建0-1规划模型;类似于顶点距离,将某些汽配件的颜色或类别不相邻要求定义为0-1矩阵,巧妙地构造了喷涂生产的约束条件.该建模方法简单快速,通用性高,适用于具有类似特点的各类生产实践问题.     

Optimisation of the ultrasound-assisted extraction of betalains and polyphenols from Amaranthus hypochondriacus var. Nutrisol

The present study optimised the ultrasound-assisted extraction (UAE) of bioactive compounds from Amaranthus hypochondriacus var. Nutrisol. Influence of temperature (25.86–54.14 °C) and ultrasonic power densities (UPD) (76.01–273.99 mW/mL) on total betalains (BT), betacyanins (BC), betaxanthins (BX), total polyphenols (TP), antioxidant activity (AA), colour parameters (L*, a*, and b*), amaranthine (A), and isoamaranthine (IA) were evaluated using response surface methodology. Moreover, betalain extraction kinetics and mass transfer coefficients (K_La) were determined for each experimental condition. BT, BC, BX, TP, AA, b*, K_La, and A were significantly affected (p < 0.05) by temperature extraction and UPD, whereas L*, a*, and IA were only affected (p < 0.05) by temperature. All response models were significantly validated with regression coefficients (R²) ranging from 87.46 to 99.29%. BT, A, IA, and K_La in UAE were 1.38, 1.65, 1.50, and 29.93 times higher than determined using conventional extraction, respectively. Optimal UAE conditions were obtained at 41.80 °C and 188.84 mW/mL using the desired function methodology. Under these conditions, the experimental values for BC, BX, BT, TP, AA, L*, a*, b*, K_La, A, and IA were closely related to the predicted values, indicating the suitability of the developed quadratic models. This study proposes a simple and efficient UAE method to obtain betalains and polyphenols with high antioxidant activity, which can be used in several applications within the food industry.     

A self-powered and sensitive terahertz photodetection based on PdSe2

With the rapid development of terahertz technology, terahertz detectors are expected to play a key role in diverse areas such as homeland security and imaging, materials diagnostics, biology, medical sciences, and communication. Whereas self-powered, rapid response, and room temperature terahertz photodetectors are confronted with huge challenges. Here, we report a novel rapid response and self-powered terahertz photothermoelectronic (PTE) photodetector based on a low-dimensional material: palladium selenide (PdSe₂). An order of magnitude performance enhancement was observed in photodetection based on PdSe₂/graphene heterojunction that resulted from the integration of graphene and enhanced the Seebeck effect. Under 0.1-THz and 0.3-THz irradiations, the device displays a stable and repeatable photoresponse at room temperature without bias. Furthermore, rapid rise (5.0 μs) and decay (5.4 μs) times are recorded under 0.1-THz irradiation. Our results demonstrate the promising prospect of the detector based on PdSe₂ in terms of air-stable, suitable sensitivity and speed, which may have great application in terahertz detection.     

Optimization of heterogeneous Catalyst-assisted fatty acid methyl esters biodiesel production from Soybean oil with different Machine learning methods

There is a growing attention to the bio and renewable energies due to fast depletion of fossil fuels as well as the global warming problem. Here, we developed a modeling and simulation method by means of artificial intelligence (AI) for prediction of the bioenergy production from vegetable bean oil. AI methods are well known for prediction of complex and nonlinear process. Three distinct Adaptive Boosted models including Huber regression, LASSO, and Support Vector Regression (SVR) as well as artificial neural network (ANN) were applied in this study to predict actual yield of Fatty acid methyl esters (FAME) production. All boosted utilizing the Adaptive boosting algorithm. The important influencing parameters on the biodiesel production such as the catalyst loading (CAO/Ag, wt%) and methanol to oil (Soybean oil) molar ratio were selected as the input variables of models while the yield of FAME production was selected as output. Model hyper-parameters were tuned to maintain generality while improving prediction accuracy. The models were evaluated using three distinct metrics Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and R². Error rates of 8.16780E-01, 4.43895E-01, 2.06692E + 00, and 3.92713 E-01 were obtained with the MAE metric for boosted Huber, SVR, LASSO and ANN models. On the other hand, the RMSE error of these models were about 1.092E-02, 1.015E-02, 2.669E-02, and 1.01174E-02, respectively. Finally, the R-square score were calculated for boosted Huber, boosted SVR, and boosted LASSO as 0.976, 0.990, 0.872, and 0.99702, respectively. Therefore, it can be concluded that although the boosted SVR and ANN models were better models for prediction of process efficiency in terms of error, but all algorithms had high accuracy. The optimum yield of 83.77% and 81.60% for biodiesel production were observed at optimum operating values from boosted SVR and ANN models, respectively.     

Design,synthesis and evaluation of novel dehydroabietic acid-dithiocarbamate hybrids as potential multi-targeted compounds for tumor cytotoxicity

In the present study, novel representatives of the important group of biologically-active, dehydroabietic acid-bearing dithiocarbamate moiety, were synthesized and characterized by ¹H NMR, ¹³C NMR, HR-MS. The in vitro antiproliferative activity evaluation (MTT) indicated that these compounds exhibited potent inhibitory activities in various cancer cell lines (HepG-2, MCF-7, HeLa, T-24, MGC-803). Particularly, compound III-b possessed extraordinary cytotoxicity with low micromolar IC₅₀ values ranging from 4.07 to 38.84 µM against tested cancer cell lines, while displayed weak cytotoxicity on two normal cell lines (LO-2 and HEK 293 T). Subsequently, the potential mechanisms of representative compound III-b were elementarily investigated by Transwell experiment, which showed III-b can inhibit cancer cells migration. Annexin-V/PI dual staining showed that the compound can induce HepG-2 cells apoptosis in a dose-dependent manner. Meanwhile this apoptosis may be related to the upregulated protein expression of cleaved-caspase 3, cleaved-caspase 9, Bax and downregulated of Bcl-2 indicated by Western Blot. Later study further confirmed that ROS levels in HepG-2 cells increased significantly with the rise of concentrations. In addition, through the network pharmacology data analyzing, the core targets and signaling pathways of compound III-b for treatment of liver neoplasms were forecasted. Molecular docking model showed that compound III-b had high affinity with hub targets (CASP3, EGFR, HSP90AA1, MAPK1, ERBB2, MDM2), suggesting that compound III-b might target the hub protein to modulate signaling activity. Taken together, these data indicated that dehydroabietic acid structural modification following the “Molecular hybridization” principle is a feasible way to discover the potential multi-targeted antitumor compounds.     

Theoretical study of (e,2e) triple differential cross sections of pyrimidine and tetrahydrofurfuryl alcohol molecules using multi-center distorted-wave method

We report theoretical studies of electron impact triple differential cross sections of two bio-molecules,pyrimidine and tetrahydrofurfuryl alcohol,in the coplanar asymmetric kinematic conditions with the impact energy of 250 eV and ejected electron energy of 20 eV at three scattering angles of-5°,-10°,and-15°.Present multi-center distorted-wave method well describes the experimental data,which was obtained by performing(e,2e)experiment.The calculations show that the secondary electron produced by the primary impact electron is strongly influenced by the molecular ionic multi-center potential,which must be considered when the low energy electron interacts with DNA analogues.