基于改进MobileFaceNet的人脸识别方法北大核心

针对目前大多数人脸识别算法参数多、计算量大,难以部署到移动端和嵌入式设备中的问题,提出了一种基于改进MobileFaceNet的人脸识别方法。通过对MobileFaceNet模型结构的调整,将bottleneck模块优化为sandglass模块,改良深度卷积和逐点卷积的相对位置,适当增大sandglass模块的输出通道数,从而减少特征压缩时的信息丢失,增强人脸空间特征的提取。实验结果表明:改进后的方法在LFW测试数据集上准确率达99.15%,模型大小和计算量分别仅为原算法的61%和45%,验证了所提方法的有效性。     

Mutual information maximizing GAN inversion for real face with identity preservation

Recent generative adversarial networks (GANs) have yielded remarkable performance in face image synthesis. GAN inversion embeds an image into the latent space of a pretrained generator, enabling it to be used for real face manipulation. However, current inversion approaches for real faces suffer the dilemma of initialization collapse and identity loss. In this paper, we propose a hierarchical GAN inversion for real faces with identity preservation based on mutual information maximization. We first use a facial domain guaranteed initialization to avoid the initialization collapse. Furthermore, we prove that maximizing the mutual information between inverted faces and their identities is equivalent to minimizing the distance between identity features from inverted and original faces. Optimization for real face inversion with identity preservation is implemented on this mutual information-maximizing constraint. Extensive experimental results show that our approach outperforms state-of-the-art solutions for inverting and editing real faces, particularly in terms of face identity preservation.     

Multi-scale and multi-patch transformer for sandstorm image enhancement

Sandstorm is a meteorological phenomenon common in arid and semi-arid regions. A sandstorm can carry large volumes of sand unexpectedly, which leads to severe color deviations and significantly degraded visibility when an image is taken in such a scenario. However, existing image enhancement methods cannot enhance sandstorm images well due to the challenging degradations and the scarcity of sandstorm training data. In this paper, we propose a Transformer with rotary position embedding to perform sandstorm image enhancement via building multi-scale and multi-patch dependencies. Our key insights in this work are 1) a multi-scale Transformer can globally eliminate the color deviations of sandstorm images via aggregating global information, 2) a multi-patch Transformer can recover local details well via learning the spatial variant degradations, and 3) a U-shape Transformer with rotary position embedding as the core unit of multi-scale and multi-patch Transformer can effectively build the long-range dependencies. We also contribute a real-world Sandstorm Image Enhancement (SIE) dataset including 1,400 sandstorm images with different degrees of degradations and various scenes. Experiments performed on synthetic images and real-world sandstorm images demonstrate that our proposed method not only obtains visually pleasing results but also outperforms state-of-the-art methods qualitatively and quantitatively.     

碳中和背景下配电网差异化节能降耗潜力评估分

为了提高配电网差异化节能降耗效果，解决现有潜力评估方法存在的应用性能差的问题，提出碳中和背景下配电网差异化节能降耗潜力优化评估方法。根据配电网的空间结构，构建相应的等值电路模型。在该模型下，从设备损耗和运行附加损耗2个方面计算配电网的损耗量。根据损耗量计算结果，确定配电网差异化碳中和节能降耗方式。从静态和动态2个角度设置潜力评估指标，通过指标数据处理、指标权重求解等步骤，得出配电网差异化节能降耗潜力的综合量化评估结果。将设计潜力评估方法应用到配电网的差异化节能降耗改造工作中，能够有效降低配电网的实际线损量、降低区域损耗费用，并具有较高的应用价值。     

高深溜井井筒堵塞的爆破处理实践

溜井是采用平硐-溜井方式开拓矿山的运输咽喉,溜井正常与否对矿山生产影响极大。本文通过黑沟铁矿高深溜井井筒堵塞处理实例,对其堵塞爆破处理方法及经验做了系统分析。重点介绍的爆破冲击波破拱疏通高深溜井井筒高位堵塞的爆破方法,富有新意,可供国内外同类型矿山参考借鉴。     

Microbial fuel cell (MFC)-based biosensor for combined heavy metals monitoring and associated bioelectrochemical process

To explore the feasibility and related mechanism of MFC biosensor for wastewater detection under the action of combined heavy metals. Cyclic voltammetry (CV) and scanning electrochemical impedance spectroscopy (EIS) were used to explore the related bioelectrochemical process. The response of the reactor to single/combined heavy metals, low/high heavy metal concentrations, and the differences in ohmic resistance (Rs) and charge transfer resistance (Rct) were investigated using Ni as the core heavy metal and the combined action of Cd, Cu and Zn. The results indicated that there was a linear relationship between the concentration and output voltage of the MFC biosensor under the action of combined heavy metals (R² = 0.8803–0.973). However, the internal resistance (Rint) of the MFC biosensor under the action of single heavy metal was far less than that of the combined heavy metal group, and the power density (19.849 W m^?3) was 4 times that of the combined heavy metal group (3.109–4.589 W m^?3). The Rs of the biosensors in the combined heavy metal group were 0.868Ω and 0.860, which were higher than 0.768Ω of the single heavy metal sensor. With the increase of the concentration of heavy metals in the influent, the increase of Rct was more obvious in the combined group, while the Rs in the single group significantly increased (P < 0.05). The results imply that it is possible for MFC biosensors to be used in the detection of actual water polluted by various heavy metals, but the biosensor performance is mainly limited by Rct, which needs to be further improved.     

塔里木盆地塔西南坳陷柯东构造带甫沙4井原油来源及充注过程

甫沙4井位于塔里木盆地塔西南坳陷昆仑山前冲断带的柯东构造带上，北部和东部分别发育有柯克亚和柯东1井油气田。为研究甫沙4井原油来源与充注过程，对原油样品和连续抽提后的含油砂样各组分（游离态、束缚态、包裹体）进行GC、GC?MS和 GC?IRMS分析，与柯克亚凝析油气田油样进行油—油对比。结果表明：甫沙4井晚期充注原油组分具有C_29?32重排藿烷、重排甾烷和Ts相对含量高，C_27?29甾烷ααα 20R分布呈反“L”型，以及正构烷烃单体碳同位素值较低等特征，与柯克亚凝析油气田来源于二叠系普司格组（P_2?3p）烃源岩的主体原油（I类）地球化学特征一致。而早期充注的原油组分具有重排藿烷、重排甾烷和Ts相对含量较低，C_27?29甾烷ααα 20R分布呈“V”型，以及正构烷烃单体碳同位素值较高等特征，与柯克亚凝析油气田来源于中—下侏罗统湖相泥岩的II类原油地球化学特征一致。甫沙4井经历3个阶段成藏过程：①在上新世，二叠系烃源岩于生油晚期阶段生成的I类原油运移至柯克亚构造带或柯东构造带深部形成油藏；②在更新世早期，侏罗系烃源岩于生油早—中期生成的II类原油运移至甫沙4井白垩系储层；③在第四纪，强烈的构造作用使深部I类原油沿断裂调整进入甫沙4井白垩系储层。最终造成甫沙4井白垩系储层II类原油先充注，I类原油后充注的特殊现象。     

“半翻转”教学模式下评价指标体系研究与实践

研究基于“半翻转”教学模式的线上教学评价指标体系,包括课前活动开展、课中教学活动设计、课后任务以及基于此过程中的评价指标体系设计。最后通过层次分析法设计评价指标权重,设计评价算法得到动态评价结果。该评价指标体系应用在本专业“数据结构与算法”课程中,并得到良好的应用效果。     

疫情下电磁场理论课程的网络教学

本文主要总结了新冠疫情期间作者的电磁场理论课程在线教学经验。对比分析了录播和直播的优缺点后,选择录播教学方式。基于超星网络教学平台,展示了录播网络教学的具体措施,包括网上答疑和学习效果检查以及在线批改作业等。给出了网络教学可以为线下教学继续使用的方法和手段,为疫情结束后的正常教学提供了新的网络教学补充措施。     

5G核心网业务模型的智能化预测研究

摘 要：核心网业务模型的建立是5G网络容量规划和网络建设的基础，通过现有方法得到的理论业务模型是静态不可变的且与实际网络存在偏离。为了克服现有5G核心网业务模型与现网模型适配性较差以及规划设备无法满足用户实际业务需求的问题，提出了一种长短期记忆（long short-term memory，LSTM）网络与卷积LSTM （convolution LSTM，ConvLSTM）网络双通道融合的 5G 核心网业务模型预测方法。该方法基于人工智能（artificial intelligence，AI）技术以实现高质量的核心网业务模型的智能预测，形成数据反馈闭环，实现网络自优化调整，助力网络智能化建设。