Line-based camera calibration with lens distortion correction from a single image

Camera calibration is a fundamental and important step in many machine vision applications. For some practical situations, computing camera parameters from merely a single image is becoming increasingly feasible and significant. However, the existing single view based calibration methods have various disadvantages such as ignoring lens distortion, requiring some prior knowledge or special calibration environment, and so on. To address these issues, we propose a line-based camera calibration method with lens distortion correction from a single image using three squares with unknown length. Initially, the radial distortion coefficients are obtained through a non-linear optimization process which is isolated from the pin-hole model calibration, and the detected distorted lines of all the squares are corrected simultaneously. Subsequently, the corresponding lines used for homography estimation are normalized to avoid the specific unstable case, and the intrinsic parameters are calculated from the sufficient restrictions provided by vectors of homography matrix. To evaluate the performance of the proposed method, both simulative and real experiments have been carried out and the results show that the proposed method is robust under general conditions and it achieves comparable measurement accuracy in contrast with the traditional multiple view based calibration method using 2D chessboard target.     

Schematic eye with a gradient-index lens and aspheric surfaces

A new schematic eye with aspheric surfaces and a radially varying refractive-index distribution lens is proposed. Image quality and spherical aberration are determined by use of ray tracing, and the results are presented as spot diagrams and compared with five existing model eyes. The proposed model provides the best image quality and lowest spherical aberration.     

Aberration correction using a multi-segment mirror with feedback interferometry

A novel real-time aberration correction system using feedback interferometry and a multi-segment mirror for the wavefront corrector is described. The result is a simple and high-speed adaptive optics system demonstrating that feedback interferometry may potentially be an alternative method to conventional aberration correction techniques. Its high-speed processing and simplicity makes real-time correction practical. Our results show that the system can respond up to 37 Hz for sinusoidal phase variations with amplitude of one wavelength and improves Strehl ratio significantly.     

一种薄非球面透镜的加工与检测

介绍了外径为 φ14 8mm、厚度为 5mm的月牙形非球面透镜加工的一些工艺难点及解决方法 ,并给出了检验方案。对加工过程中出现的变形原因及克服方法进行了探讨 ,并给出了透镜最终的检测结果。     

Assessing impact force localization by using a particle swarm optimization algorithm

This paper focuses on the inverse problem regarding force localization in the case of impacts not concentrated at a point but which occur on elastic beams. Following the identification approach proposed to solve this problem and which is based on the reciprocity theorem, the impact location characteristics were determined by using particle swarm optimization algorithm. To eliminate numerical trouble due to the trivial solutions appearing in this formulation, the fitness function was customized by introducing a set of weighting coefficients. Four different formulations of the fitness function were considered and their performances with regards to the number of sensors used and their positions were analyzed. They enabled a selection of the best combination of weighting coefficients to be used in the context of an impact force localization process based on the particle swarm optimization technique. Three sensors were found to be required and comparison with a genetic algorithm has revealed the effectiveness of the proposed method in terms of accuracy and computational time.     

基于粒子群算法的稀疏阵列超声相控阵全聚焦成像

为降低相控阵超声检测全聚焦算法的成像数据量及阵列稀疏优化的计算时间,研究了一种用于稀疏阵列全聚焦成像的阵列优化算法,并通过实验对其成像效果进行了验证。针对目前超声相控阵检测的全矩阵采集数据量大、全聚焦算法成像时间长的难点,该文通过构建稀疏阵列,在保证成像质量的同时显著降低成像数据量,提高了全聚焦算法的成像效率。通过以主瓣宽度、旁瓣峰值以及主瓣峰值作为约束条件构建适应度函数,采用粒子群算法得到稀疏阵元位置分布并进行阵元权重修正,并将其用于稀疏全聚焦成像。相比全阵元成像,使用粒子群算法所得的稀疏阵列的阵元个数降低了56.25%、65.62%,数据使用量降低了80.86%、88.18%。在阵列优化方面,相比遗传算法减少了84.86%的计算时间。     

轴对称非球面透镜光轴共轴度的测量研究

介绍了一种轴对称非球面透镜的光轴共轴度的测量方法。激光管发出的光束经束腰变换透镜入射到被测透镜的非球面表面,由CCD摄像头接收非球面的反射激光光斑,CCD的光敏面位于反射激光光束的束腰位置;调整被测透镜位置,直到激光束腰中心位置不随被测非球面透镜的旋转而变化,这说明被测透镜的非球面对称轴与机械旋转轴重合;再利用球面偏心测量原理检测被测透镜球面一面的偏心量,即可以求得被测非球面透镜的光轴共轴度。该测量方法的误差小于20″。该方法适用于判定非球面透镜和非球面反射镜是否合格,以及调整非球面透镜的制造工艺。     

Shape measurement of aspheric plastic lens with large angle

In this paper, according to the features of easy distortion and scratch for aspheric plastic lens, a noncontact measuring method is raised to test error in shape of the lens. Namely, the distance between a template and its image reflected with tested lens can be measured in nearly the vertical direction of the lens axis when the two-dimensional (2D) template is put near the measured surface. Then, the outline of the central cross-section could be obtained by calculating and curve fitting. Furthermore, three-dimensional (3D) surface can be imitated through rotating the component. A new fitting method of drift measurement is presented to prevent reducing precision when the lens and the template are fixed. The template is adjusted according to the position of the lens. The measurement precision is in the order of magnitued of sub-microns. Rotationally symmetric convex aspheric surface with any angle can be measured by this method.     

透镜偏振像差分析

本文对轴上物点发出的光束经透镜后的偏振像差进行了定量分析。     

Designing a plano-convex aspheric lens for fiber optics collimator

A plano-convex aspheric lens is designed to collimate the emitted light of the fiber optics. In this paper even orders of general aspheres are used to describe the aspheric surfaces sag. To determine the aspheric coefficient, the even asphere polynomial are fitted on the computed sag data. The surface sag data is determined using genetic algorithm method. The surface sag was analyzed by the commercial optical software package ZEMAX. For typical fiber optics, the numerical aperture (NA) of 0.22 was assumed. With this specification, the expected output spot diagram of 20 mm and the maximum aperture of 25.4 mm have been obtained. A comparison between here presented lens and the similar type of conventional lens has been carried out; the proposed aspheric lens corrected the spherical aberration, which led to increase in the collimated distance.     

微动粒子群优化算法用于Egun的多参量优化

针对2维电子光学多参量优化问题,采用微动粒子群优化算法,在给出目标电子轨迹和优化范围的前提下,可以得到趋近于该电子轨迹的真空边界和聚束磁结构。该算法分为前后两阶段：第一阶段采用前后试探法(微动),同时参照最优粒子的信息;第二阶段采用标准粒子群优化算法。针对涉及多个相关参量的电子光学设计问题,标准粒子群优化算法仅能保证以较高概率收敛到局部最佳解,而微动粒子群优化算法能以较高概率收敛到全局最佳解,并且展现了多核计算机在电子光学设计上的潜力。初步的软件试验显示：消耗人类工程师几周时间的电子光学设计问题,用微动粒子群算法在普通个人计算机上几十小时就能完成。     

High-pressure structures of InBi predicted by particle swarm optimization algorithm

We extensively explore the high-pressure structures of InBi by using a newly developed particle swarm optimization algorithm. An orthorhombic Imma structure is discovered to be stable from 43.7 GPa to 107.9 GPa, ruling out the previously speculated cubic structure. Further increasing the pressure, we find a tetragonal P4/nmm structure which is energetically more favourable from 107.9 GPa to 200 GPa. Especially, the tetragonal P4/nmm structure is known to occur at high pressure in the structures of ZnO and MgTe. We also predict this structure to be a high-pressure structure of ZnTe. Thus the tetragonal P4/nmm structure may be a universal high-pressure structure of the Ⅱ-Ⅵ and the Ⅲ-Ⅴ compounds.     

基于粒子群优化的混沌系统比例-积分-微分控制

基于比例-积分-微分(PID)控制算法的简单性和实用性,但对于复杂非线性系统控制时参数的难以确定问题,运用集群智能中的改进粒子群算法进行PID控制器的优化,并应用于若干混沌系统的控制.对Hénon混沌、Duffing混沌、六辊UC 轧机混沌、Nagumo-sato神经元混沌、Chen氏混沌以及永磁同步电动机混沌的控制进行了仿真研究.研究结果表明： 用PID进行混沌系统的输出反馈控制是有效的,从而拓宽了PID控制的应用范围; 用简单方法控制复杂混沌系统是完全可能的,对混沌系统的控制具有较好的参考价值; 粒子 关键词： 混沌 比例-积分-微分控制 粒子群优化算法     

Aberration improvement of gradient index rod lens

The aberration of gradient index (GRIN) rod lenses induced by deviation from the ideal refractive index distribution is decreased by optimized thermal diffusion. The index distribution coefficients are estimated by a novel method. The lenses' heat treatment process and aberration testing system are described.     

Modeling of pedestrian evacuation based on the particle swarm optimization algorithm

By applying the evolutionary algorithm of Particle Swarm Optimization (PSO), we have developed a new pedestrian evacuation model. In the new model, we first introduce the local pedestrian’s density concept which is defined as the number of pedestrians distributed in a certain area divided by the area. Both the maximum velocity and the size of a particle (pedestrian) are supposed to be functions of the local density. An attempt to account for the impact consequence between pedestrians is also made by introducing a threshold of injury into the model. The updating rule of the model possesses heterogeneous spatial and temporal characteristics. Numerical examples demonstrate that the model is capable of simulating the typical features of evacuation captured by CA (Cellular Automata) based models. As contrast to CA-based simulations, in which the velocity (via step size) of a pedestrian in each time step is a constant value and limited in several directions, the new model is more flexible in describing pedestrians’ velocities since they are not limited in discrete values and directions according to the new updating rule.     

双压电片变形镜像差校正能力

利用实测影响函数,通过数值仿真分析了有效孔径、光束入射角度对于20单元双压电片变形镜像差校正能力的影响。仿真结果表明：有效孔径为16 mm时,变形镜对于各阶像差的校正能力较好;而随着光束入射到变形镜角度的增大,变形镜对于像差的校正精度与校正幅度都会出现下降,但是如果能够将入射角控制在25以内,入射角度对变形镜的像差校正能力影响不大。     

双压电片变形镜像差校正能力

利用实测影响函数,通过数值仿真分析了有效孔径、光束入射角度对于20单元双压电片变形镜像差校正能力的影响。仿真结果表明:有效孔径为16 mm时,变形镜对于各阶像差的校正能力较好;而随着光束入射到变形镜角度的增大,变形镜对于像差的校正精度与校正幅度都会出现下降,但是如果能够将入射角控制在25以内,入射角度对变形镜的像差校正能力影响不大。     

2维静磁场多参量微动粒子群优化算法

提出一种微动粒子群优化算法,针对2维静磁场多参量优化问题,在给出轴上目标轴向磁感应强度分布曲线的前提下,可以得到趋近于该分布曲线的磁结构。该算法分为前后两阶段:第一阶段采用前后试探法(微动),同时也参照最优粒子的信息;第二阶段采用基本粒子群优化算法。微动粒子群优化算法可以发挥多核计算机在工程设计上的潜力,而且即使粒子数目很少,也能不断趋近目标解。     

An experiment for obtaining DOP ellipsoid using particle swarm optimization algorithm

The degree of polarization (DOP) ellipsoid can be used as either feedback or feedforward signal for automatic polarization mode dispersion compensation. We have realized the experiment for obtaining DOP ellipsoid from 100 sampling data of output states of polarization using particle swarm optimization (PSO) as ellipsoid data fitting algorithm. It was shown that the PSO algorithm was powerful for ellipsoid data fitting with high precision within 250 ms.