基于相对梯度的人脸识别方法

在原始相对梯度算子的基础上,提出一种新的相对梯度算子,并将它与二维主成分分析（2DPCA）或者二维Fisher线性判别分析（2DFLD）相结合,形成一种基于改进相对梯度算子的人脸识别算法。在AR库和Yale_B库上的实验表明,基于改进相对梯度算子的人脸识别算法对人脸图像的光照、表情等变化均具有较好的鲁棒性,识别准确率明显高于只用2DPCA或2DFLD进行特征抽取的人脸识别方法,以及基于原始相对梯度算子的人脸识别算法。同时采用三种不同大小的窗口分别进行实验,实验结果证明,当窗口大小为3×3时,识别效果相对最好。     

急倾斜水平分段放顶煤放煤规律的离散元模拟研究

提高水平分段高度是提高急倾斜水平分段放顶煤产量的有效途径,但顶煤分段高度提高后,其顶煤的放出工艺如何控制是影响其放出率的关键因素。对急倾斜水平分段放顶煤的放煤过程进行了离散元数值模拟,重点比较了在相同顶煤高度的条件下,顺序放煤与隔架放煤的特点,并在理论分析中给出了确定放煤轮数的具体公式,所得结论可供工程实际参考应用。     

抽放方法治理采煤工作面上隅角瓦斯

针对高瓦斯工作面上隅角瓦斯大,严重影响安全,制约生产的问题,采用直接抽放工作面上隅角瓦斯及顶板抽放法取得成功。实践证明顶板抽放法是解决高瓦斯采面瓦斯问题非常有效和更为安全的方法。     

基于Modbus现场总线的综采工作面大型设备监测监控系统

简要介绍了Modbus通讯协议,重点阐述了综采工作面大型设备实现远程监测监控功能的软硬件技术方案。在HT6L1-400Z/1140智能型组合电器的基础上采用Modbus现场总线,编制了通讯程序,实现了综采工作面大型设备的远程监测监控。     

π放面机道用凝胶防灭火的实践

鹤岗矿业集团公司大陆煤矿在-275南11层零区回掘期间,应用新型凝胶防灭火工艺,成功地完成了回采任务,有效地消灭了煤层发火,保证了正常生产。     

高瓦斯自燃煤层综放工作面瓦斯综合治理

通过对高瓦斯自燃煤层综采放顶煤回采工作面的瓦斯涌出考察,及对瓦斯涌出与风量、高抽巷布置、推进速度、周期来压的关系分析,提出瓦斯综合治理措施,在实践中取得较好成效。     

Constructing Facial Identity Surfaces for Recognition

We present a novel approach to face recognition by constructing facial identity structures across views and over time, referred to as identity surfaces, in a Kernel Discriminant Analysis (KDA) feature space. This approach is aimed at addressing three challenging problems in face recognition: modelling faces across multiple views, extracting non-linear discriminatory features, and recognising faces over time. First, a multi-view face model is designed which can be automatically fitted to face images and sequences to extract the normalised facial texture patterns. This model is capable of dealing with faces with large pose variation. Second, KDA is developed to compute the most significant non-linear basis vectors with the intention of maximising the between-class variance and minimising the within-class variance. We applied KDA to the problem of multi-view face recognition, and a significant improvement has been achieved in reliability and accuracy. Third, identity surfaces are constructed in a pose-parameterised discriminatory feature space. Dynamic face recognition is then performed by matching the object trajectory computed from a video input and model trajectories constructed on the identity surfaces. These two types of trajectories encode the spatio-temporal dynamics of moving faces.     

基于复小波变换全监督学习的微震波形降噪与相位拾取

由于井下微震信号有着更低的信噪比, 导致信号拾取精度降低. 现阶段基于小波阈值的信号降噪算法等在面对信噪比较低的信号时存在泛化性差, 阈值难以衡量等问题. 为解决这一问题, 本文研究了一种复小波变换全监督学习的微震波形降噪方法. 该方法首先利用复小波变换结合卷积自编器设计一个具有多个卷积和反卷积操作的编码-解码器完成图像的降噪过程. 为验证此方法的有效性, 首先在Stanford的Earthquake数据集上构建了Earthquake2023进行训练和测试, 并有着较好地拟合效果和训练结果. 同时基于该方法降噪后信号设计了一种震相拾取方法, 并达到了较高的拾取精度. 本文设计了多组对比实验, 结果表明此降噪方法能有效提高信号的峰值信噪比和均方根误差, 两者分别提高了16 dB和24%, P波、S波初至到时拾取的误差相较于STA/LTA减小了0.3 ms.     

A Fully-Human Antibody Specifically Targeting a Membrane-Bound Fragment of CADM1 Potentiates the T Cell-Mediated Death of Human Small-Cell Lung Cancer Cells

Small-cell lung cancer (SCLC) is the most aggressive form of lung cancer and the leading cause of global cancer-related mortality. Despite the earlier identification of membrane-proximal cleavage of cell adhesion molecule 1 (CADM1) in cancers, the role of the membrane-bound fragment of CAMD1 (MF-CADM1) is yet to be clearly identified. In this study, we first isolated MF-CADM1-specific fully human single-chain variable fragments (scFvs) from the human synthetic scFv antibody library using the phage display technology. Following the selected scFv conversion to human immunoglobulin G1 (IgG1) scFv-Fc antibodies (K103.1–4), multiple characterization studies, including antibody cross-species reactivity, purity, production yield, and binding affinity, were verified. Finally, via intensive in vitro efficacy and toxicity evaluation studies, we identified K103.3 as a lead antibody that potently promotes the death of human SCLC cell lines, including NCI-H69, NCI-H146, and NCI-H187, by activated Jurkat T cells without severe endothelial toxicity. Taken together, these findings suggest that antibody-based targeting of MF-CADM1 may be an effective strategy to potentiate T cell-mediated SCLC death, and MF-CADM1 may be a novel potential therapeutic target in SCLC for antibody therapy.     

IUINet: 基于Shift的双流映射3D医学分割模型

为了提高特征融合, 我们设计了动态全连接层(DyFC), 该方法重新定义了权重和偏置, 使用基向量来代表新的权重和偏置, 基向量的系数是根据每一个输入特征进行学习得到的, 权重和偏置不再是共享的, 而是特有的, 这对于每一个特征的表达更具有专向性. 在本文中, 我们提出了一种双流映射结构模型IUINet. IUINet是通过3DShift操作、空间可分离卷积的组合来实现医学图像分割任务, 同时保持精度和效率之间的平衡. 所提出来的IUINet遵循编码器-解码器结构, 其中编码器一部分包含Shift操作、逐点Conv1×1操作, 另一部分包含空间可分离卷积操作. IUINet运用了多尺度输入以及多尺度特征映射层, 提高反向传播速度, 降低反向传播的平均距离. 提高模型的精确度, 增加模型泛化能力, 减少过拟合.