浅渍法发酵豇豆的工艺优化

以具有降解亚硝酸盐功能的植物乳杆菌（Lactobacillus plantarum）SD-7和具有优良抗氧化能力的植物乳杆菌（Lactobacillus plantarum）FM-LP-9为复合发酵剂（1∶1），采用浅渍法发酵豇豆。以硬度和感官评分为考察指标，通过单因素试验及响应面试验研究复合发酵剂接种量、发酵温度和发酵时间对浅渍法发酵豇豆品质的影响。结果表明，最佳发酵工艺条件为发酵温度25 ℃，接种量5%，发酵时间125 h，在此优化条件下，得到的浅渍法发酵豇豆硬度为47.31 N，感官评分为90.98分，香气浓郁、口感脆嫩。     

Foam-based cleaning of surfaces contaminated with mixtures of oil and soot

Liquid foams of intermediate stability have been shown to be very efficient in the cleaning of sensitive surfaces because of the synergy between imbibition and foam decay. While we quantified these mechanisms for contaminations with liquid oils in our previous work, we show here their extension to oils containing soot particles in an effort to simulate increasingly realistic contaminations. Using foams with a wide range of liquid fractions and with different stabilities, we show that the main cleaning mechanisms remain very similar, with the oil entraining the soot particles. However, we find much less efficient soot removal when the liquid channels of the foams are small enough to hinder efficient transport of the soot particles.     

Computerized morphometry analysis of epithelial fimbriae nuclear symmetry in BRCA carriers may identify patients at risk for developing ovarian cancer

Serous ovarian tumors may originate in epithelial cells of the fallopian tubes. Computerized morphometry was able to find significant alterations in the fallopian tube epithelium of healthy BRCA carriers. The purpose of this study was to identify a subgroup of BRCA carriers that may be at risk of developing serous ovarian cancer by evaluation of the epithelial nuclear symmetry in the fallopian tubes. Four groups of patients were analyzed; healthy patients, ovarian cancer patients, BRCA carriers, and BRCA noncarriers. All fallopian tubes appeared normal by H&E examination. The ImageProPlus software was used to assess the nuclear symmetry of 65 fimbriae epithelium cells and an artificial neural network algorithm aided in detecting a subpopulation among fimbriae of healthy BRCA carriers at risk for ovarian cancer. Significant differences were found between healthy patients and ovarian cancer patients, and between BRCA carriers and noncarriers. The algorithm was able to accurately predict BRCA carriers with associated ovarian cancer based on fallopian tube nuclear symmetry characteristics. These results reinforce the hypothesis that fimbriae epithelial cells of BRCA carriers may undergo early-stage changes that could predict the risk of progression toward malignancy.     

High-roubustness keystroke recognition method based on acoustic spatial gradient

For the fluctuation of CFCC caused by environmental noise is the main reason for the low accuracy of keystroke detection,the spatial characteristics of adjacent between CFCC were studied,and the spatial gradient structure of CFCC based on points was established.On this basis,the effect of CFCC spatial gradient on keystroke content recognition and the selection of precise neighborhood points were studied on training and testing.Finally,a high-robustness keystroke recognition algorithm based on acoustic signals was constructed.Extensive experiments in different environments demonstrate that the proposed CFCC spatial gradient sound feature achieves great performance and the recognition accuracy is 96.15%.     

结合注意力的3D卷积网络脑胶质瘤分割算法

为了提升脑胶质瘤分割精度，提出一种结合注意力机制的3D卷积神经网络算法。输入3个不同尺度的图像块，经过9个卷积层和1个分类层后得到3个不同的分类结果，将分类结果与注意力学习到的权重相乘并逐体素相加得到输出。此外该算法采用了一种混合Dice损失函数与Focal损失函数的超参数损失函数。实验表明，该算法的Dice系数在整体区域、核心区域以及增强区域分别达到了95.31%、80.12%、82.25%。与已有的一种脑胶质瘤分割算法deepmedic相比，整体区域、核心区域以及增强区域的Dice系数分别提升了3%、2%、6%。在脑胶质瘤分割方面，具有重要的临床意义。     

基于卷积神经网络的单目深度估计

针对利用深度学习方法对街道图像进行深度估计，提出采用语义分割的方法解决深度图出现边界模糊等问题；估计深度通过左右视角图生成视差图进行无监督的训练。在网络模型中添加语义分割层，采取多个空洞卷积并行的结构增加感受野，同时减少了图像下采样的次数，降低了由于下采样带来的信息损失，使得的结果更加准确。这也是在深度估计中首次与空洞卷积相结合增加准确率。通过对KITTI街道数据集进行训练，与现有结果相比，除了增加检测准确性，降低错误率之外，使得效果图中的物体更加清晰，并且在效果图中还保留了一些原模型中被忽视掉的细节信息，将原始图像更加完整的表现出来。     

Passive target detection and tracking from electromagnetic field measurements

This paper presents a novel approach to the localization of moving targets in a complex environment based on the measurement of the perturbations induced by the target presence on an independently‐generated time‐varying electromagnetic field. Field perturbations are measured via a set of sensors deployed over the domain of interest and used to detect and track a possible target by resorting to a particle Bernoulli filter (PBF). To comply with real‐time operation, the PBF works along with an artificial neural network (ANN) model of the environment trained offline via finite elements (FEs). The performance of the proposed algorithm is assessed via simulation experiments.     

基于改进CGANs的入侵检测方法研究

近年来,机器学习算法在入侵检测系统(IDS)中的应用获得越来越多的关注。然而,传统的机器学习算法更多的依赖于已知样本,因此需要尽可能多的数据样本来对模型进行训练。遗憾地是,随着越来越多未知攻击的出现,且用于训练的攻击样本具有不平衡性,传统的机器学习模型会遇到瓶颈。文章提出一种将改进后的条件生成对抗网络(CGANs)与深度神经网络(DNN)相结合的入侵检测模型(CGANs-DNN),通过解决样本不平衡性问题来提高检测模型对未知攻击类型或只有少数攻击样本类型的检测率。深度神经网络(DNN)具有表征数据潜在特征的能力,而经过改进后的条件CGANs,能够通过学习已知攻击样本潜在数据特征分布,来根据指定类型生成新的攻击样本。此外,与生成对抗网络(GANs)和变分自编码器(VAE)等无监督生成模型相比,CGANsDNN经过改进后加入梯度惩罚项,在训练的稳定性上有了很大地提升。通过NSL-KDD数据集对模型进行评估,与传统算法相比CGANs-DNN不仅在整体准确率、召回率和误报率等方面有更好的性能,而且对未知攻击和只有少数样本的攻击类型具有较高的检测率。     

视频侦查中人像智能分析应用及算法优化

人像智能分析指的是对视频或录像中的人像进行结构化和可视化分析,对目标人物进行性别、年龄、发型等特征的智能识别,这项技术在视频侦查中有极高的应用价值。人像识别早期的算法是通过人工提取特征,通过学习低级视觉特征来针对不同属性进行分类学习,这种基于传统方法的模型表现常常不尽如人意。在计算机视觉领域,通过海量图像数据学习的神经网络比传统方法有更丰富的信息量和特征可以被提取。文章尝试通过深度学习技术训练神经网络模型对行人进行检测和识别,对于衣着不同的行人进行智能识别,具有更好的鲁棒性,提升了视频人像识别的准确率,拓展了人工智能技术在身份识别领域的应用。     

基于模拟退火优化BP神经网络的变形监测数据预测方法

针对BP神经网络模型用于变形监测数据处理时容易陷入局部极小值并且收敛速度慢的问题，提出一种基于模拟退火(Simulated Annealing,SA)算法优化BP神经网络的建筑物变形预测方法，利用SA的全局寻优能力对BP神经网络的模型参数进行优化，使参数迭代过程始终保持较高的“温度”和“能量”，从而确保BP神经网络能够得到全局最优解的同时具备较高的预测精度和收敛速度。采用实际算例对所提SA-BP方法在低信噪比和小样本条件下的预测精度进行验证，结果表明所提方法相对于传统BP神经网络法和小波方法能够获得更高的预测精度，并且在小样本和低信噪比条件下优势更加明显。