基于旋转描述的三维几何造型方法

提出了一种能用凸四边形表示并可作平面展开的三维实体造型的设计方法。该方法用旋转描述表把二维平面展开图和三维实体联系起来，模拟人工折叠过程，由二维平面展开图逐次旋转变换完成三维造型重建。造型重建后，再通过纹理映射将平面设计图案映射到实体的各个面上，完成对实体着色渲染，生成具有真实感的三维实体。     

机器人头部的视觉跟踪系统研究

对运动目标进行识别和跟踪是机器人视觉的一项关键技术。提出了一种可以快速跟踪运动目标的系统设计方案,采用了双计算机处理和Memolink通信方式的机器人控制结构,以图像处理为基础,在图像经过校正之后,通过区域分割和目标识别两个步骤来确定目标的位置,并控制机器人对目标进行跟踪。实验表明,该方法具有较高的速度和准确性。     

面阵探测器的像点亚像素定位研究

面阵探测器可直接探测点目标图像的位置，但其定位精度受到探测器分辨率的限制。因此，亚像元精度技术得到了发展。它是利用目标像点的灰度分布特性，通过内插细分算法，确定出像点位置。可将定位精度提高1—2个数量级。本文介绍了几种实现目标像点亚像素定位的内插细分算法。对它们的计算公式、特点进行了分析，并进行了实验验证。     

测定透镜厚度新法及其误差探讨

本文介绍利用景深短、高倍率、大口径的物镜，采用光学方法，对被测透镜上下表面调焦，从而测定其中心厚度的一种新方法，并对测量误差进行了探讨。     

Effective electromagnetic and chiral parameters of chiral composite material

The effective parameters of chiral composite are studied using a simple model, that is, randomly oriented non-interacting wire helices embedded in a nonchiral host medium. It is found that both the effective permittivity and permeability are independent on the handedness of the chiral objects while the effective chirality admittance is dependent. It is also found that when the ratio of the radius of the chiral helix to its pitch is about 0.23, maximum chirality admittance is achieved. The effective parameters of equichiral sample are also discussed.     

Bistatic scattering by arbitrarily shaped objects with rough surface at optical and infrared frequencies

The scattering cross sections for arbitrarily shaped dielectric objects with rough surface are determined for optical and infrared frequencies using the Kirchhoff approximation. The formula of the coherent scattering cross section is derived, and numerical method of incoherent scattering cross section is given. As a specific example, the infrared laser scattering cross sections of rough spheres are calculated at 1.06 m.     

Scale Enhancement Pyramid Network for Small Object Detection from UAV Images

Object detection is challenging in large-scale images captured by unmanned aerial vehicles (UAVs), especially when detecting small objects with significant scale variation. Most solutions employ the fusion of different scale features by building multi-scale feature pyramids to ensure that the detail and semantic information are abundant. Although feature fusion benefits object detection, it still requires the long-range dependencies information necessary for small objects with significant scale variation detection. We propose a simple yet effective scale enhancement pyramid network (SEPNet) to address these problems. A SEPNet consists of a context enhancement module (CEM) and feature alignment module (FAM). Technically, the CEM combines multi-scale atrous convolution and multi-branch grouped convolution to model global relationships. Additionally, it enhances object feature representation, preventing features with lost spatial information from flowing into the feature pyramid network (FPN). The FAM adaptively learns offsets of pixels to preserve feature consistency. The FAM aims to adjust the location of sampling points in the convolutional kernel, effectively alleviating information conflict caused by the fusion of adjacent features. Results indicate that the SEPNet achieves an AP score of 18.9% on VisDrone, which is 7.1% higher than the AP score of state-of-the-art detectors RetinaNet achieves an AP score of 81.5% on PASCAL VOC.     

圆柱浮体的平衡姿态及其稳定性

浮心相当于浮体的支承点,其位置随物体的姿态而变化;物体稳定平衡时重心与浮心的距离达到极小。圆柱浮体有2~5个平衡位置,其中相间的1~2个稳定,其姿态与比重和长径比相关,在系统参数变化时可能突变。     

Unsupervised cell identification on multidimensional X‐ray fluorescence datasets

A novel approach to locate, identify and refine positions and whole areas of cell structures based on elemental contents measured by X‐ray fluorescence microscopy is introduced. It is shown that, by initializing with only a handful of prototypical cell regions, this approach can obtain consistent identification of whole cells, even when cells are overlapping, without training by explicit annotation. It is robust both to different measurements on the same sample and to different initializations. This effort provides a versatile framework to identify targeted cellular structures from datasets too complex for manual analysis, like most X‐ray fluorescence microscopy data. Possible future extensions are also discussed.     

Coupled Navier–Stokes/molecular dynamics simulations in nonperiodic domains based on particle forcing

This paper focuses on coupling methods for hybrid Navier–Stokes/molecular dynamics (MD) simulations. The computational domain is split in a continuum flow region, where a finite‐volume discretisation of the Navier–Stokes equations is used, and one or more particle domains, where molecular level modelling of the flow is employed. The domains are defined with a partial overlap, in which the flow states are coupled through an exchange of the velocity components. For the steady flows considered, an under‐relaxed Newton iteration method is used to drive the coupled system to convergence. The main focus of the present work is on methods to impose nonperiodic boundary conditions on the particle domain(s). A particle forcing is applied in the direction normal to the particle domain boundary to impose the boundary normal velocity component. A novel aspect of the present work is the extension of this method to more general nonplanar particle domain boundaries. The main contribution of the paper is the development of a particle forcing method in the direction tangential to the domain boundary, which is based on the equivalent continuum‐flow boundary shear stresses along with an iterative forcing strength adjustment based on the extrapolated particle boundary velocity. Furthermore, an adaptation scheme is presented, which uses the finite‐volume flux residuals of the particle bin averaged velocity field as a truncation criterion for the iterative force‐update scheme. It is demonstrated that by comparing the residual reduction for the momentum equation in the nonhomogeneous directions during the molecular dynamics simulations with that for a homogeneous direction, the forcing iteration at which the statistical noise in the velocity field dominates the uncertainty in the forcing strength can be determined. At this point the iteration can be truncated. It is shown that with adaptive schemes of this type, the total number of MD evaluations required in a coupled Navier–Stokes/MD simulation can be reduced relative to a hybrid scheme with a fixed number of forcing‐strength updates. Copyright © 2011 John Wiley & Sons, Ltd.