基于阶段注意力机制的电力线提取算法

电力系统维护是电力系统稳定运行的重要保障，应用智能算法的无人机电力巡检则为电力系统维护提供便捷。电力线提取是自主电力巡检以及保障飞行器低空飞行安全的关键技术，结合深度学习理论进行电力线提取是电力巡检的重要突破点。本文将深度学习方法用于电力线提取任务，结合电力线图像特点嵌入改进的图像输入策略和注意力模块，提出一种基于阶段注意力机制的电力线提取模型（SA-Unet）。本文提出的SA-Unet模型编码阶段采用阶段输入融合策略（Stage input fusion strategy， SIFS），充分利用图像的多尺度信息减少空间位置信息丢失。解码阶段通过嵌入阶段注意力模块（Stage attention module，SAM）聚焦电力线特征，从大量信息中快速筛选出高价值信息。实验结果表明，该方法在复杂背景的多场景中具有良好的性能。     

基于Bert预训练模型的虚假新闻文本检测

虚假新闻剥夺了人们获取真相的权利,也给社会和国家带来了许多危害。文中以虚假新闻文本为例,分析和验证了多种预训练语言模型在虚假新闻文本分类上的分类效果。经实验证明,相较于其他语言模型,Bert预训练语言模型取得了较好的结果,预测准确率为86.97%,召回率为88.12%,F1值为87.54%。     

A hybrid knowledge and ensemble classification approach for prediction of venous thromboembolism

Clinical narratives such as progress summaries, lab reports, surgical reports, and other narrative texts contain key biomarkers about a patient's health. Evidence-based preventive medicine needs accurate semantic and sentiment analysis to extract and classify medical features as the input to appropriate machine learning classifiers. However, the traditional approach of using single classifiers is limited by the need for dimensionality reduction techniques, statistical feature correlation, a faster learning rate, and the lack of consideration of the semantic relations among features. Hence, extracting semantic and sentiment-based features from clinical text and combining multiple classifiers to create an ensemble intelligent system overcomes many limitations and provides a more robust prediction outcome. The selection of an appropriate approach and its interparameter dependency becomes key for the success of the ensemble method. This paper proposes a hybrid knowledge and ensemble learning framework for prediction of venous thromboembolism (VTE) diagnosis consisting of the following components: a VTE ontology, semantic extraction and sentiment assessment of risk factor framework, and an ensemble classifier. Therefore, a component-based analysis approach was adopted for evaluation using a data set of 250 clinical narratives where knowledge and ensemble achieved the following results with and without semantic extraction and sentiment assessment of risk factor, respectively: a precision of 81.8% and 62.9%, a recall of 81.8% and 57.6%, an F measure of 81.8% and 53.8%, and a receiving operating characteristic of 80.1% and 58.5% in identifying cases of VTE.     

基于深度学习的秀丽隐杆线虫显微图像分割方法

利用计算机实现自动、准确的秀丽隐杆线虫(C.elegans)的各项形态学参数分析,至关重要的是从显微图像上分割出线虫体态,但由于显微镜下的图像噪声较多,线虫边缘像素与周围环境相似,而且线虫的体态具有鞭毛和其他附着物需要分离,多方面因素导致设计一个鲁棒性的C.elegans分割算法仍然面临着挑战。针对这些问题,提出了一种基于深度学习的线虫分割方法,通过训练掩模区域卷积神经网络(Mask R-CNN)学习线虫形态特征实现自动分割。首先,通过改进多级特征池化将高级语义特征与低级边缘特征融合,结合大幅度软最大损失(LMSL)损失算法改进损失计算;然后,改进非极大值抑制;最后,引入全连接融合分支等方法对分割结果进行进一步优化。实验结果表明,相比原始的Mask R-CNN,该方法平均精确率(AP)提升了4.3个百分点,平均交并比(mIOU)提升了4个百分点。表明所提出的深度学习分割方法能够有效提高分割准确率,在显微图像中更加精确地分割出线虫体。     

A Student's t-based density peaks clustering with superpixel segmentation (tDPCSS) method for image color clustering

Image color clustering is a basic technique in image processing and computer vision, which is often applied in image segmentation, color transfer, contrast enhancement, object detection, skin color capture, and so forth. Various clustering algorithms have been employed for image color clustering in recent years. However, most of the algorithms require a large amount of memory or a predetermined number of clusters. In addition, some of the existing algorithms are sensitive to the parameter configurations. In order to tackle the above problems, we propose an image color clustering method named Student's t-based density peaks clustering with superpixel segmentation (tDPCSS), which can automatically obtain clustering results, without requiring a large amount of memory, and is not dependent on the parameters of the algorithm or the number of clusters. In tDPCSS, superpixels are obtained based on automatic and constrained simple non-iterative clustering, to automatically decrease the image data volume. A Student's t kernel function and a cluster center selection method are adopted to eliminate the dependence of the density peak clustering on parameters and the number of clusters, respectively. The experiments undertaken in this study confirmed that the proposed approach outperforms k-means, fuzzy c-means, mean-shift clustering, and density peak clustering with superpixel segmentation in the accuracy of the cluster centers and the validity of the clustering results.     

基于机器视觉的运煤车车号识别技术研究

在铁路运煤装车过程中为了快速、准确地识别车号,提出一种基于机器视觉的运煤车车号识别技术。将连通区域提取与投影分割法结合,实现车号的粗定位、细分割,并对图像中的断裂字符进行二次分割,构建了基于BP神经网络的分类模型进行车号识别,提升了煤炭装车的效率和精度。     

基于微博用户兴趣话题的相似用户挖掘

相似用户挖掘是提高社交网络服务质量的重要途径，在面向大数据的社交网络时代，准确的相似用户挖掘对于用户和互联网企业等都有重要的意义，而根据用户自己的兴趣话题挖掘的相似用户更符合相似用户的要求。提出了一种基于用户兴趣话题进行相似用户挖掘的方法。该方法首先使用TextRank话题提取方法对用户进行兴趣话题提取，再对用户发表内容进行训练，计算出所有词之间的相似度。提出CP（Corresponding Position similarity）、CPW（Corresponding Position Weighted similarity）、AP（All Position similarity）、APW（All Position Weighted similarity）四种用户兴趣话题词相似度计算方法，通过用户和相似用户间关注、粉丝重合率验证相似用户挖掘效果，APW similarity的相似用户的关注/粉丝重合百分比为1.687%，优于提出的其他三种算法，分别提高了26.3%、2.8%、12.4%，并且比传统的文本相似度方法Jaccard相似度、编辑距离算法、余弦相似度分别提高了20.4%、21.2%、45.0%。因此APW方法可以更加有效地挖掘出用户的相似用户。     

基于查询的新闻多文档自动摘要技术研究

针对新闻文本领域,该文提出一种基于查询的自动文本摘要技术,更加有针对性地满足用户信息需求。根据句子的TF-IDF、与查询句的相似度等要素,计算句子权重,并根据句子指示的时间给定不同的时序权重系数,使得最近发生的新闻内容具有更高的权重,最后使用最大边界相关的方法选择摘要句。通过与基于TF-IDF、Text-Rank、LDA等六种方法的对比,该摘要方法ROUGE评测指标上优于其他方法。从结合评测结果及摘要示例可以看出,该文提出的方法可以有效地从新闻文档集中摘取核心信息,满足用户查询内容的信息需求。     

结合位图切割和区域合并的彩色图像分割

就经典分水岭图像分割算法中存在的过分割问题，提出一种结合位图切割和区域合并的彩色图像分割算法。对原始彩色图像通过空域梯度算子求其梯度图像，并利用位图切割重建梯度图像；对新梯度图像进行分水岭预分割；对预分割图像基于异质性最小原则进行区域合并，并获得最终分割结果。相比于现有的同类方法，该算法引入位图切割，抑制噪声对分割结果的影响，在边缘模糊处分割准确，得到符合人类视觉的较小分割区域数目，同时在运行效率上提高。