基于多阶融合与循环聚合的立体匹配网络

针对基于深度学习的立体匹配网络中病态区域匹配效果欠佳、模型参数量过大的问题,提出了一种基于多阶特征融合与循环代价聚合的端对端立体匹配网络—MFRANet。首先,为兼顾图像低层细节信息与高层语义信息,提出了多阶特征融合模块,采用分阶段、逐步式的特征融合策略对多层次、多尺度特征进行有效融合;其次,在代价聚合阶段提出循环聚合机制,以循环方式对匹配代价卷进行聚合优化,在改善聚合效果的同时不引入过多的参数量;最后,利用基于Soft Argmin算法的视差计算模块计算图像视差。并通过KITTI 2012/2015和SceneFlow两个公开数据集对网络进行训练和测试,与其他端对端立体匹配网络进行了对比研究。结果表明,在SceneFlow和KITTI 2015两个公开数据集上,相较于其他端对端立体匹配网络,MFRANet具有更为精准的匹配结果;对于SceneFlow数据集,终点误差降低至0.92Pixels;对于KITTI 2015数据集,误匹配率降低至2.21%。

面向部分向量化的循环分布及聚合优化

大量循环中都存在着少数无法向量化的语句以及许多可向量化语句,循环分布通常可以将这些语句分离到不同的循环中,进而实现循环的部分向量化。目前主流的优化编译器仅支持简单激进的循环分布方法,因而导致向量化后的循环开销过大,且不利于寄存器和cache的重用。针对上述问题,提出了面向部分向量化的循环分布及聚合方法。首先,分析了一般循环分布的两个关键问题:语句集的划分和循环执行顺序的确定;其次,提出了面向最大聚合的凝聚图结点排序方法来指导循环合并,在不影响并行性的前提下减小了循环开销;最后,通过实验对提出的方法进行了验证。实验结果表明,对于测试用例,提出的方法能够生成正确的向量化代码,并且能够显著提高向量化程序的执行效率。

聚合循环泵机械密封泄漏原因分析及改进

聚丙烯装置因聚合循环泵机械密封泄漏多次出现非计划停车,影响了安全生产。通过分析机械密封结构、计算端面比压,找到了机械密封泄漏的原因。对大气侧机械密封动环材质及其端面的结构形式进行改进后,有效解决了聚合循环泵机械密封泄漏问题。

基于循环特征融合的弯道增强车道线检测算法

针对道路转弯处曲率过大导致弯道识别的精度下降的问题,本文提出了一种基于循环特征融合Resa-CC的弯道增强车道线检测算法。该算法利用车道线的形状先验性,捕获图像像素中行与列的空间关系,融合信息生成特征图;以残差网络为主体框架,加入编码器、解码器和注意力机制模块,在损失函数中引入弯道结构约束来提高车道线弯道的识别精度。加入循环特征融合模块和自注意力机制模块后准确率分别提升3.41%和1.1%,证明了两模块的有效性;Resa-CC算法准确率可达96.83%,FPS为35.68,误检率FP和漏检率FN分别为0.0315和0.0282,表明本文算法具有较高的检测性能,在车辆行驶弯道路段中能更准确地推断出车道线的位置。

《聚合物资源循环利用技术》“金课”建设的探索与实践

聚合物材料的广泛应用带来了严重的环境问题和资源问题,因此废旧聚合物资源循环利用十分必要。依托长期在聚合物资源循环利用领域的科研积累,我们针对资源循环科学与工程专业开设了《聚合物资源循环利用技术》课程。本课程以“金课”的高阶性、创新性和挑战度为标准,建立了《聚合物资源循环利用技术》的教学内容、教学方法和考核制度,以培养学生运用基础理论知识解决复杂实际问题的综合能力和科研创新能力。

废旧聚合物材料的化学循环利用

本文在介绍高分子废料循环方法之后，着重综述各种聚合物材料化学循环的状态及其发展，并描述了化学循环的工艺及其设备，最后对化学循环的发展趋势和前景进行讨论。

基于改进VariFocalNet的微小目标检测

针对航拍场景中包含的目标尺寸小、有效特征信息少的问题,提出一种基于改进的变焦网络VFNet(VariFocalNet)的航拍场景中微小目标检测算法。首先,为增强微小目标的特征表征能力,采用特征提取性能更好的循环层聚合网络(RLANet)代替ResNet作为主干网络;其次,为解决特征金字塔自顶向下融合时顶层特征信息丢失问题,引入特征增强模块(FEM);然后,为解决现有标签分配方法在微小目标标签分配上的样本分布不平衡问题,改进的VFNet采用了基于高斯感受野的标签分配方法;最后,为减小微小目标对位置偏移的敏感性,引入一种边界框回归损失函数Wasserstein损失测量预测边界框高斯分布和真值框高斯分布的相似性。在AI-TOD数据集上的实验结果表明:改进后的VFNet算法的平均精度均值(mAP)达到了14.9%;与改进前的算法相比,在航拍场景下的微小目标上的检测mAP提高了4.7个百分点。

多尺度特征多径自适应复用的显著性目标检测

由于忽略了对多尺度特征的提取,以及不同层次特征之间的差异,显著性目标检测仍然存在预测结构不完整、细节丢失的问题。为了解决这些问题,提出了一个新的显著性检测模型M3Net。该网络主要由多尺度特征自适应融合模块和循环反馈聚合模块组成。多尺度特征自适应融合模块旨在自适应地捕捉和聚合不同层次的多尺度特征。循环反馈聚合模块组成可以在迭代过程中对不同层次特征聚合的同时,有效地防止特征的稀释。在五个基准数据集上的实验结果表明,该网络在F_(β)、E_(m)、MAE三种评价指标上优于十种现有网络。在DUT-OMRON数据集上,F_(β)指标比排名第二的显著性检测模型提高了0.4%,E_(m)指标提高了0.3%;在ECSSD数据集上,Fβ指标比排名第二的显著性检测模型提高了0.2%,E m指标提高了0.3%,同时网络还具有优秀的速度表现。

碳中和目标背景下包装行业发展的对策

目的为了应对日益恶化的全球气候变化,并实现碳中和的目标,研究低碳化技术在包装行业的应用成为当下至关重要的任务。为了推动包装行业在碳中和背景下的低碳化转型,有必要对现有的低碳化技术进行分类和梳理。方法从碳减排和碳移除两方面对包装行业已有的低碳化技术进行了研究。综述当前可降解生物塑料和聚合物单体化学循环技术在包装中的应用,介绍了几种清洁能源以及碳捕捉、碳利用和碳储存技术的发展现状。结论对包装行业而言,实现碳中和的目标还面临着很多挑战。包装行业在推进碳中和目标时要选取与发展适合的低碳能源和碳中和技术。

基于改进VarifocalNet的高精度目标检测算法

针对通用目标检测场景下,现有单阶段无锚检测器识别精度低、识别困难等问题,提出一种基于改进变焦网络VFNet(VarifocalNet)的高精度目标检测算法。首先,利用循环层聚合网络(RLANet)替换VFNet用于特征提取的主干网络ResNet,循环残差连接操作将前层特征汇入后续网络层中提升特征的表征能力;其次,通过带有特征对齐卷积操作的特征金字塔网络(FPN)替换原始的特征融合网络,利用可变形卷积操作在FPN上下层融合过程中实现特征对齐并优化特征表征能力;最后,使用聚焦-全局蒸馏(FGD)算法进一步提升小规模算法的检测性能。在COCO(Common Objects in Context)2017数据集上进行的评估实验结果表明,在相同训练条件下,改进后的以RLANet-50为主干的算法的均值平均精度(mAP)可以达到45.9%,与VFNet算法相比提升了4.3个百分点,而改进后的算法参数量为36.67×10^(6),与VFNet相比仅高了4×10^(6)。可见,改进后的VFNet算法在提升检测精度的同时稍微增加了参数量,说明该算法可以满足目标检测的轻量化及高精度需求。

基于YOLOv3-SE-RE模型的羊姿态识别

为了实现养殖场环境下羊只个体的有效识别及对羊只姿态进行迅速判断,针对现有模型识别精度差,效率低等问题,基于自建数据集,提出基于YOLOv3的改进模型。在主干网络darknet53中,首先引入通道注意力模块压缩激励网络,增强网络对重要通道的关注度,提高网络检测精度;其次,将网络中的部分残差模块用循环特征移位聚合模块替代,以提高检测速度和检测精度;采用余弦退火动态学习率替代原有学习率,在训练过程中进行动态微调,使网络能轻松跳出局部最优解。实验结果表明:在羊只检测与姿态识别任务中,YOLOv3-SE-RE算法的平均精度(mean Average Precision,mAP)比原始YOLOv3算法的mAP提升了9.98%,同时检测速度也得到提升。

基于ET-GMPHD算法的编队目标跟踪方法

针对低检测环境下编队目标的跟踪问题,提出了一种基于扩展目标高斯混合概率假设密度(extended target Gaussian mix probability hypothesis density,ET-GMPHD)算法的编队目标跟踪方法。首先,保留修剪掉高斯项的同时将其外推,用Jensen-Shannon(JS)散度衡量下一时刻状态估计值与外推值间的相似程度,并以此反映是否有目标丢失,保证真实目标不被修剪,解决了因目标漏检导致跟踪结果不准确的问题。其次,结合循环阈值聚合法得到编队整体的状态估计,消除了估计状态集合中状态值过多造成的影响。最后,仿真实验表明,该方法能够在检测概率极低的情况下进行有效跟踪,并具有良好的跟踪性能。

Rapid,sensitive detection of Vibrio anguillarum using loop-mediated isothermal amplification

Vibrio anguillarum is an important bacterial pathogen of aquatic organisms and a significant problem in aquatic farming. The rapid detection and identification of V. anguillarum, and other pathogens that infect marine organisms, is crucial to effective disease management. In this study, we developed a loop-mediated amplification （LAMP） assay to detect V. anguillarum in an hour in a single tube without the need for thermal cycling. Conserved regions of the metalloproteinase （empA） gene of V. anguillarum served as the targets for primer design. A fragment of the empA gene was amplified at 65℃ in the presence of the primer mixture and Bst DNA polymerase. In the optimized LAMP assay, 6.7 pg of V. anguillarum DNA could be detected. Six strains of V. anguillarum and 17 strains of non-V, anguillarum bacteria were used in this study to evaluate the species specificity of the primers. The six V. anguillarum strains gave a positive result in the LAMP assay. This method was also validated in V. anguillarum-infected fish. This LAMP method is more sensitive than PCR in the detection of V. anguillarum and shows good species specificity. The LAMP assay is therefore an effective method for the quick detection of V. anguillarum both in the laboratory and in the field.

Evaluation of substrates for zinc negative electrode in acid PbO_2–Zn single flow batteries

An investigation was performed on the suitability of carbon materials, metallic lead and its alloys as substrates for zinc negative electrode in acid PbO2-Zn single flow batteries. The zinc deposition process was carried out in the mediumofl mol.L 1H2SO4 at room temperature. No maximum current appears on the potentiostatic current transients for the zinc deposition on lead and its alloys. With increasing overpotential, the progressive nucleation turns to be a 3D-instantaneous nucleation process for the resin-graphite composite. Hydrogen evolution on the graphite composite is effectively suppressed with the doping of a polymer resin. The hydrogen evolution reaction on the lead is relatively weak, while on the lead alloys, it becomes serious to a certain degree. Although the ex- change current density of zinc deposition and dissolution process on the graphite composite is relatively low, the zinc corrosion is weakened to a great extent. With the increase of deposition time, zinc deposits are more compact. The cyclings of zinc galvanostatic charge-discharge on the graphite composite provide more than 90% of coulombic and 80% of energy efficiencies, and exhibit superior cycling stability during the first 10 cycles.

CS-2-G型催化剂在SPHERIZONE工艺上的应用

在中沙(天津)石化有限公司450 kt/a Spherizone工艺聚丙烯装置上,用CS-2-G/ZN1l8混合催化剂替代ZN118催化剂,以生产牌号为EP300H聚丙烯为例,考察装置聚合物粉料筛分、细粉含量、聚合物性能、物耗能耗。并结合研究催化剂的微观结构从理论上及实际应用考核该催化剂在该装置上的适用性。结果表明,在装置工况不变的情况下,CS-2-G催化剂可以满足生产工艺要求,可以部分替代ZN118催化剂。

Electropolymerization of O-Phenylenediamine in an Ionic Liquid

Ionic liquid like 1-ethyl-3-methylimidazolium bromine ([EMIM]Br) has been used as electrolyte for the electropolymerization of O-phenylenediamine at glassy carbon electrode by cyclic voltammetry. It is found that poly (O-phenylenediamine) film modified electrode has favorable electrochemical activity in acid solution.

Electrochemical synthesis of polyaniline in reverse microemulsion

The electro-polymerization behavior of aniline in reverse(W/O) microemulsion was investigated. The experiment results show that the cyclic voltammetry polymerization behavior of aniline in W/O microemulsion is different from that in aqueous solution remarkably. With the increase of scan cycle, the oxidation potential shifts positively and the reduction potential shifts negatively, i.e., the redox potential difference increases. H+ apparent concentration affects the aniline polymerization evidently. When H+ concentration is lower than 0.08 mol/L, the electro-polymerization of aniline is difficult. With the increase of H+ concentration, the polymerization current of aniline increases gradually. Only when H+ concentration is high enough(0.5 mol/L), aniline can be well electro-polymerized. Moreover, under the same condition, the aniline polymerization current in W/O microemulsion is higher than that in aqueous solution. The scanning electron microscopy image shows that the deposited polyaniline(PANI) has uniform fiber morphology with diameter of about 100 nm. Further study result suggests that the electrochemical activity of the PANI in HCl is similar to that of the PANI prepared in aqueous solution.

Effect of Freeze-Thaw on Water Stability of Aggregates in a Black Soil of Northeast China

Soil aggregate stability,an important index of the physical characteristics of a soil,can provide a good indication of a soil’s erodibility,and deserves special consideration in regions with cold climate.The objective of this study was to study the effect of freeze-thaw on soil water-stable aggregates in the black soil region of Northeast China.Samples of a typical black soil in the region were collected to measure water-stable aggregates after freeze-thaw under different conditions(i.e.,initial moisture contents,freezethaw cycles and freezing temperatures)by wet-sieving into eight particle size groups(>10,10–6,6–5,5–3,3–2,2–1,1–0.5,and0.5–0.25 mm).Freeze-thaw had the most effect on aggregate stability when the samples had an initial moisture content of 400 g kg-1.The water-stable aggregates of the four larger particle size groups(>5,5–3,3–2,and 2–1 mm)reached a peak stability value,but those of the two smaller particle size groups(1–0.5 and 0.5–0.25 mm)reached a minimum value when the soil moisture content was 400 g kg-1.Water-stable aggregates of the four larger particle size groups decreased while those of the two smaller particle size groups increased with the increase of freeze-thaw cycles.As temperatures fell,the water-stable aggregates of the four larger particle size groups decreased while those of the two smaller particle size groups increased.

Polyaniline-modified cetyltrimethylammonium bromide- graphene oxide-sulfur nanocomposites with enhanced performance for lithium-sulfur batteries

Conductive polymer coatings can boost the power storage capacity of lithiumsulfur batteries. We report here on the design and preparation--by combining a facile and green chemical deposition method with an oxidative polymerization approach--of polyaniline （PANI）-modified cetyltrimethylammonium bromide （CTAB）-graphene oxide （GO）-sulfur （S） nanocomposites with significantly enhanced performance in lithium-sulfur batteries. Such conductive polymer modified CTAB-GO-S nanocomposites as sulfur cathode materials can deliver high specific discharge capacities and long-term cycling performance, i.e., -970 mAh-g-1 at 0.2 C and -715 mAh-g-1 after 300 cycles, -820 mAh.g-1 at 0.5 C and -670 mAh.g-1 after 500 cycles, -770 mAh.K at 1 C and -570 mAh.g-~ after 500 cycles. The capacity decay was as low as 0.036% per cycle at 0.5 C, and 0.051% per cycle at 1 C. Under the same condition, batteries using PANI-modified CTAB-GO-S as cathodes exhibited higher specific capacity and higher average coulombic efficiency compared with CTAB-decorated GO-S and GO--S nano- composites. The improved performance can be attributed to the lower charge transfer resistance and the alleviated dissolution of polysulfides in the PANI- modified CTAB-GO-S cathodes.