RETRACTED ARTICLE: Intelligent traffic monitoring and traffic diagnosis analysis based on neural network algorithm

Traffic sign recognition and lane detection play an important role in traffic flow planning, avoiding traffic accidents, and alleviating traffic chaos. At present, the traffic intelligent recognition rate still needs to be improved. In view of this, based on the neural network algorithm, this study constructs an intelligent transportation system based on neural network algorithm, and combines machine vision technology to carry out intelligent monitoring and intelligent diagnosis of traffic system. In addition, this study discusses in detail the core of the monitoring system: multi-target tracking algorithm, and introduces the complete implementation process and details of the system, and highlights the implementation and tracking effect of the multi-target tracker. Finally, this study uses case identification to analyze the effectiveness of the algorithm proposed by this paper. The research results show that the proposed method has certain practical effects and can be used as a reference for subsequent system construction.

     

纳米载体铜基催化剂对低碳醇合成的影响

本文用沉积-沉淀法制备了新型纳米载体低碳醇合成催化剂，并与共沉淀法所制备的催化剂进行了对比，结果表明，纳米载体催化剂活性高、耐热性能好、醇收率高纳米载体对纳米载体催化剂活性大小的影响顺序为：纳米Al2O3＞纳米ZnO-ZnAl2O4＞纳米MgAl2O4＞纳米MgO.工艺参数对反应性能的影响结果表明，反应温度低，甲醇选择性高；但温度太高（＞320℃）CO2。选择性升高，且整个反应为动力学控制；压力升高可提高活性和C1-C2OH选择性，但C4OH选择性下降；结合空速对产物分布的影响，反应可能按Smith和Anderson的缩聚机理进行．     

基于BP神经网络锅炉炉膛火焰可视化监测方法研究

为了可视化监测炉膛火焰燃烧状况,提出一种基于级联前向BP神经网络模型的锅炉炉膛火焰可视化监测方法。通过比较选取级联前向BP神经网络作为炉膛温度预测模型,利用图像处理技术得到炉膛火焰辐射能图像对应二维温度场,并采用正则化方法重建炉膛火焰三维温度场。仿真结果表明,根据二维温度场可得火焰等温线走向和分布,根据三维温度场易得火焰中心分布及全炉最高温度点信息,实现锅炉运行控制和异常温度报警,满足燃烧诊断要求并实现炉膛火焰可视化监测。     

集成模糊LSA与MIL的图像分类算法

针对自然图像的分类问题,提出一种基于模糊潜在语义分析(LSA)与直推式支持向量机(TSVM)相结合的半监督多示例学习(MIL)算法.该算法将图像当作多示例包,分割区域的底层视觉特征当作包中的示例.为了将MIL问题转化成单示例问题进行求解,首先利用K-Means方法对训练包中所有的示例进行聚类,建立"视觉词汇表";然后根据"视觉字"与示例之间的距离定义模糊隶属度函数,建立模糊"词-文档"矩阵,再采用LSA方法获得多示例包(图像)的模糊潜在语义模型,并通过该模型将每个多示例包转化成单个样本;采用半监督的TSVM训练分类器,以利用未标注图像来提高分类精度.基于Corel图像库的对比实验结果表明,与传统的LSA方法相比,模糊LSA的分类准确率提高了5.6%,且性能优于其他分类方法.     

基于谱残差和多分辨率分析的显著目标检测

根据人类视觉系统的特点,提出一种融合谱残差和多分辨率分析的显著目标检测方法。该方法通过在不同尺度上计算图像的亮度、颜色以及方向特征的谱残差,构建多分辨率显著性图谱序列,然后用线性插值方法将不同分辨率的特征显著图叠加得到3个特征显著图,再利用k均值聚类算法将每个特征显著图聚为两类,选择聚类中心距离最大的特征显著图作为最终的显著图,最后经过动态阈值处理获得图像的显著目标区域。基于自然图像的显著目标检测实验结果表明,该方法具有较强的稳定性和实用性,得到较为满意的检测结果。     

Saliency guided self-attention network for pedestrian attribute recognition in surveillance scenarios

Pedestrian attribute recognition is often considered as a multi-label image classification task. In order to make full use of attribute-related location information, a saliency guided sel-attention network ( SGSA-Net) was proposed to weakly supervise attribute localization, without annotations of attribute-related regions. Saliency priors were integrated into the spatial attention module ( SAM ). Meanwhile,channel-wise attention and spatial attention were introduced into the network. Moreover, a weighted binary cross-entropy loss ( WCEL) function was employed to handle the imbalance of training data. Extensive experiments on richly annotated pedestrian ( RAP) and pedestrian attribute ( PETA) datasets demonstrated that SGSA-Net outperformed other state-of-the-art methods.     

融合SIFT和MIL的红外人脸识别方法

针对红外人脸识别问题,提出一种新的基于尺度不变特征转换(SIFT)与多示例学习(MIL)相结合的算法。该算法将图像当作多示例包,SIFT描述子当作包中的示例,利用聚类的方法对训练集中的所有SIFT描述子进行聚类,建立"视觉词汇表",再根据"视觉字"在多示例训练包中出现的频率,建立"词-文档"矩阵,采用潜在语义分析(LSA)的方法获得多示例包(图像)的潜在语义特征,将MIL问题转化成标准的有监督学习问题,即在潜在语义空间用支持向量机(SVM)求解MIL问题。基于OTCBVS标准数据集的对比实验结果表明,所提算法是可行的,且识别率明显高于其他方法。     

A multifunctional ribonuclease A-conjugated carbon dot cluster nanosystem for synchronous cancer imaging and therapy

Carbon dots exhibit great potential in applications such as molecular imaging and in vivo molecular tracking. However, how to enhance fluorescence intensity of carbon dots has become a great challenge. Herein, we report for the first time a new strategy to synthesize fluorescent carbon dots (C-dots) with high quantum yields by using ribonuclease A (RNase A) as a biomolecular templating agent under microwave irradiation. The synthesized RNase A-conjugated carbon dots (RNase A@C-dots) exhibited quantum yields of 24.20%. The fluorescent color of the RNase A@C-dots can easily be adjusted by varying the microwave reaction time and microwave power. Moreover, the emission wavelength and intensity of RNase A@C-dots displayed a marked excitation wavelength-dependent character. As the excitation wavelength alters from 300 to 500 nm, the photoluminescence (PL) peak exhibits gradually redshifts from 450 to 550 nm, and the intensity reaches its maximum at an excitation wavelength of 380 nm. Its Stokes shift is about 80 nm. Notably, the PL intensity is gradually decreasing as the pH increases, almost linearly dependent, and it reaches the maximum at a pH = 2 condition; the emission peaks also show clearly a redshift, which may be caused by the high activity and perfective dispersion of RNase A in a lower pH solution. In high pH solution, RNase A tends to form RNase A warped carbon dot nanoclusters. Cell imaging confirmed that the RNase A@C-dots could enter into the cytoplasm through cell endocytosis. 3D confocal imaging and transmission electron microscopy observation confirmed partial RNase A@C-dots located inside the nucleus. MTT and real-time cell electronic sensing (RT-CES) analysis showed that the RNase A@C-dots could effectively inhibit the growth of MGC-803 cells. Intra-tumor injection test of RNase A@C-dots showed that RNase A@C-dots could be used for imaging in vivo gastric cancer cells. In conclusion, the as-prepared RNase A@C-dots are suitable for simultaneous therapy and in vivo fluorescence imaging of nude mice loaded with gastric cancer or other tumors.     

多孔壁面微通道换热性能的实验研究

微通道散热器作为一种高效散热器件,广泛应用于微电子、光电、汽车、航天国防、能源等领域。针对传统光滑微通道传热面积小、换热性能偏低、沸腾迟滞等问题,本文提出一种多孔壁面微通道结构,并采用激光直写方法实现微通道多孔壁面的高效、稳定生成。该多孔壁面微通道显著增大了换热面积、促进流体的扰动、提供大量稳定沸腾核心,从而强化单相与两相沸腾传热。通过搭建微通道换热性能测试系统,测试对比了多孔壁面微通道与光滑微通道的单相对流、两相沸腾传热性能。发现多孔壁面微通道的Nu数相对于光滑微通道提升了21%~31%。在两相沸腾换热过程中,其粗糙多孔结构促进了沸腾气泡成核,其核态沸腾起始温度相比于光滑微通道降低了35%。同时粗糙多孔结构可以保证沸腾过程中的液体持续供给,从而大幅提升了沸腾换热能力,避免了干涸现象的提前发生,其两相沸腾换热系数相对于未处理的光滑微通道最大提升了83%。此外,还开展了不同流量下多孔壁面微通道的沸腾传热性能测试,发现在质量流率为G=500kg/(m²·s)下的沸腾换热系数相对于G=200kg/(m²·s)情况下最大提升了30%。     

浅析业主对建设工程施工索赔的预防及处理

从业主的角度出发,对建设工程施工索赔管理的基本思路进行了论述,对索赔预防及处理的具体措施进行了分析,并对业主的施工索赔管理提出了几点思考。