基于改进YOLOv3的输电线路部件实时检测

针对基于深度学习的目标检测技术应用于工业领域无法在移动端嵌入式设备上实现高效且准确的检测这一问题,提出一种基于YOLOv3改进的输电线路部件实时检测算法轻量级特征融合检测模型LFF-DM(Lightweight Feature Fusion Detection Model)。一方面通过改进式的k-means算法得到聚类结果;另一方面结合深度可分离卷积和倒残差块设计出轻量化的网络结构。通过在自建的包含绝缘子、悬垂线夹、防震锤、鸟巢与导地线的专业巡检数据集上进行实验,结果表明在NVIDIA Jetson AGX Xavier设备上可以实现25FPS的检测速度及90.48%mAP的检测精度,适用于输电线路移动端实时精确巡检。     

Synthesis of polyether polyols with epoxidized soy bean oil

Epoxidized soy bean oil (ESBO) polyether polyols have been prepared and evaluated as potential bio-renewable replacements for bisphenol A based epoxy coatings. Zinc triflate was found to be more efficient in catalyzing the ESBO hydroxyl reaction than methanesulfonic acid or boron trifluoride etherate. With an excess of n-butanol, ESBO epoxide groups ring open to give the expected polyether polyol, but as the n-butanol concentration is reduced, dimers, trimers, and higher molecular weight analogs of the triglycerides appear. Weight average molecular weight can be increased in a controlled fashion to over 10,000 Da by using trimethylolpropane (TMP) in place of n-butanol. The addition of solvent reduces molecular weight of the polyether polyol, at an equivalent TMP level while still allowing good reaction control. These polyether polyols can be cured with phenolic resins, but solvent and blush resistance, adhesion, and wedge bend flexibility are inferior to a commercial bisphenol A epoxy control.     

The effect of morphology on the corrosion inhibition and mechanical properties of hybrid polymer coatings

The durability and mechanical properties of epoxy ester coatings and films has been improved by blending with rigid aromatic polyurea (PU). The interaction of PU and epoxy ester was enhanced by coupling the polymers with polymethylhydrosiloxane. The reactions between various entities are analyzed by Fourier transform infrared spectroscopy and the change in physical and mechanical properties are studied by a dynamic mechanical analyzer. The corrosion resistance of the hybrid coatings was measured by direct current polarization method, direct current polarization (DCP). The addition of polymethylhydrosiloxane enhances the corrosion properties in the hybrid coatings. The surface morphology was analyzed by scanning electron microscopy. The glass transition temperature of the films increases with increasing PU concentration and a wide glass rubber transition range for hybrid coatings was achieved which confirms the higher impact strength of the hybrid coatings and films. © 2012 Wiley Periodicals, Inc. J. Appl. Polym. Sci., 2013     

Using Knowledge-Based Neural Networks to Improve Algorithms: Refining the Chou–Fasman Algorithm for Protein Folding

This article describes a connectionist method for refining algorithms represented as generalized finite-state automata. The method translates the rule-like knowledge in an automaton into a corresponding artificial neural network, and then refines the reformulated automaton by applying backpropagation to a set of examples. This technique for translating an automaton into a network extends the KBANN algorithm, a system that translates a set of prepositional rules into a corresponding neural network. The extended system, FSKBANN, allows one to refine the large class of algorithms that can be represented as state-based processes. As a test, FSKBANN is used to improve the Chou–Fasman algorithm, a method for predicting how globular proteins fold. Empirical evidence shows that the multistrategy approach of FSKBANN leads to a statistically-significantly, more accurate solution than both the original Chou–Fasman algorithm and a neural network trained using the standard approach. Extensive statistics report the types of errors made by the Chou–Fasman algorithm, the standard neural network, and the FSKBANN network.     

Trace element analysis of blood samples from mentally challenged children by PIXE

The accelerator based ion beam analysis method of proton induced X-ray emission (PIXE) has been used for analysing up to 14 elements in the blood serum of patients, collected from rehabilitation centres for the mentally retarded and from Medical College Hospital, Coimbatore, Tamil Nadu, India. The experimental subjects of the different groups displayed significant variations in their levels of certain trace elements such as zinc, iron, copper, phosphorus, chlorine, and rubidium. The results are compared with those of healthy control subjects and are discussed in detail in this paper. Hence, PIXE as a method of trace element analysis can be used to determine trace element content in mentally challenged patients.     

Demosaicing enhancement using pixel-level fusion

Bayer pattern has been widely used in commercial digital cameras. In NASA’s mast camera (Mastcams) onboard the Mars rover Curiosity, Bayer pattern has also been used in capturing the RGB bands. It is well known that debayering, also known as demosaicing in the literature, introduces artifacts such as false colors and zipper edges. In this paper, we first present four fusion approaches, including weighted and the well-known alpha-trimmed mean filtering approaches. Each fusion approach combines demosaicing results from seven debayering algorithms in the literature, which are selected based on their performance mentioned in other survey papers and the availability of open source codes. Second, we present debayering results using two benchmark image data sets: IMAX and Kodak. It was observed that none of the seven algorithms in the literature can yield the best performance in terms of peak signal-to-noise ratio (PSNR), CIELAB score, and subjective evaluation. Although the fusion algorithms are simple, it turns out that the debayering performance can be improved quite dramatically after fusion based on our extensive evaluations. In particular, the average PSNR improvements of the weighted fusion algorithm over the best individual method are 1.1 dB for the IMAX database and 1.8 dB for the Kodak database, respectively. Third, we applied the various algorithms to 36 actual Mastcam images. Subjective evaluation indicates that the fusion algorithms still work well, but not as good as the existing debayering algorithm used by NASA.     

Electrical properties of flexible graphene reinforced polyimide composites

The electrical properties of polyimide and the composite at different volumes fractions were studied in the frequency range 200–20 kHz and in the temperature range 30–200 °C. Increasing the volume fraction of graphene up to 10%, resulted in an extremely large increase in the dielectric constant, which indicates the composites remarkable ability to store electric potential energy under the effect of alternating electric field. An increase in dielectric constant was also observed with increasing temperature and decreasing frequency. The outstanding dielectric properties of polyimide graphene nanocomposites are attributed to the large volume fraction of interfaces in the bulk of the material. The measured increase in dielectric constant with increasing temperature was attributed to the segmental mobility of the polymer chains. The AC conductivity for polyimide and the composites was calculated from the loss factor and a remarkably high conductivity was obtained for the composites due to the formation of conducting paths in the matrix by the graphene sheets. Also this study showed that the thermal conductivity of the composites increased sharply with increasing graphene concentration. © 2017 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2017 , 134, 45372.     

无人汽车的万里长征

话说在国力昌盛的汉武帝时代,中国驴友宗师张骞奉汉武帝之命两次出使西域,让人没有想到的是,他这两次公款旅游竟然走出了丝绸之路的雏形。随着东汉驴友班超再一次公款出行对线路进一步完善后,一条以长安、洛阳为起点,经甘肃、新疆,到中亚、西亚,并联结地中海各国的超远距离陆上通道得以形成,也就是著名的丝绸之路。自从丝绸之路诞生后,众多欧洲驴友抱着对物产丰富的天朝的景仰而开始了穿越欧亚之旅,其中意大利尤甚,还出了一个叫马可波罗的家伙。如今意大利人又开始重温起这丝路之旅了,只不过这一次他们玩的是新潮,万里长征干脆由无人汽车来全程代劳。     

Fusion of artificial neural networks for learning capability enhancement: Application to medical image classification

Artificial neural network (ANN) is one of the commonly used tools for computational applications. The specific advantages of ANN are high accuracy, less convergence time, less computational complexity, and so forth. However, all these merits are not available in the same ANN. Even though back propagation neural (BPN) networks are accurate, their computational complexity is significantly high. BPN networks are also not stable. On the other hand, Hopfield neural network (HNN) is better than BPN in terms of computational efficiency. But the accuracy of HNN is low. In this work, a modified ANN is proposed to overcome this specific problem. The modified ANN is a fusion of BPN and HNN. The technical concepts of BPN and HNN are mixed in the training algorithm of the proposed back propagation‐Hopfield network (BPHN). The objective of this fusion is to improve the performance of conventional ANN. Magnetic resonance brain image classification experiments are used to analyse the proposed BPHN. Experimental results have suggested improvement in the learning process of the proposed BPHN. A comparative analysis with the conventional networks is performed to validate the performance of the proposed approach.     

表层沙强度特性的试验研究

本项研究利用土壤动态三轴仪对表层沙的强度特性进行了试验分析,测出其应力应变关系和强度参数φ、c,并分析了颗粒级配、含水量、密度和动态载荷对强度特性的影响,建立了浅层沙的Duncan-Chang应力应变模型,为车辆行走机构与沙相互作用的力学计算分析提供了重要的资料。