遗传算法在圆度误差评定中的应用

针对传统最小二乘圆优化算法评定圆度误差局部收敛的问题,提出基于遗传算法的圆度误差评定方法。介绍了遗传算法基本原理和运算流程,在建立最小二乘圆数学模型的基础上,推导出待优化的目标函数,并详细描述了基于遗传算法的圆度评定优化步骤。实例计算结果表明该算法能比传统优化算法收敛到更精确的解,并具有较高的稳定性,能有效地克服局部收敛的问题。     

基于PSO-ELM算法的红外目标模拟器校准数据拟合方法研究

提高红外目标模拟器校准数据的拟合精度,对于红外目标的辐射照度等辐射特性的测量有着重要意义;针对校准数据具有很强的非线性,传统的拟合算法精度不高的问题,引入一种基于粒子群算法优化的极限学习机算法（PSO-ELM）,以标准黑体辐射温度作为输入因子,以MCT探测器实际测量出的辐射照度作为输出因子,建立PSO-ELM模型,利用粒子群算法（PSO）对连接隐藏神经元和输入层的权值和隐藏神经元阈值进行优化,拟合出输入参数和输出参数之间的非线性关系;这两个参数的优化提高了极限学习机算法（ELM）的性能,该方法的主要优点是具有较强的容错性、较好的对复杂非线性数据处理性能和ELM算法参数设置上的优化机制;通过与GA-ELM模型、ELM模型进行对,验证了与传统数据拟合方法相比,基于PSO-ELM的方法拟合精度有了很大提高,为红外目标模拟器校准数据拟合提供了新的方法。     

求解带约束的多峰函数优化问题的新演化算法

求解带约束的多峰函数优化问题在科学研究和工程应用中具有重要的现实意义。本文在对郭涛算法和“两阶段子空间演化算法”改进的基础上,通过采用混沌初始化和混沌变异算子,给出了一种求解带约束条件的多峰函数优化问题的有效的演化算法。通过对具有多峰的复杂函数的测试表明:该算法在求解质量和收敛速度方面都得到了很好的结果,表现出较强的鲁棒性。     

基于改进飞蛾扑火优化算法的MIMO雷达相位编码信号集设计

提出一种MIMO雷达相位编码信号集设计方法，该方法对飞蛾扑火优化算法进行改进，利用改进的飞蛾扑火算法优化相位编码信号集，提高其正交性.在改进方法中，增加对火焰的更新过程，使火焰不仅能够保存最优飞蛾位置还具有搜索最优解的功能.优化过程以自相关函数旁瓣峰值和、互相关函数峰值和，以及自相关函数旁瓣能量、互相关函数能量作为算法适应度函数.仿真实验结果表明，改进方法明显提高了飞蛾扑火优化算法的收敛速度和收敛精度，且优化得到的信号集具有较低的自相关旁瓣值和互相关值.     

基于Oracle罚函数的自适应约束差分进化算法

为有效求解约束优化问题,减少算法参数,提出基于Oracle罚函数方法的自适应约束差分进化算法。为满足求解优化问题的常用标准,提出一种改进的Oracle罚函数方法。将改进的Oracle罚函数方法与三种自适应差分进化算法相结合,提出三种自适应约束差分进化算法。对11个典型测试函数的优化结果验证了Oracle罚函数方法与自适应差分进化算法结合的有效性。与参考文献中提出的算法的比较结果表明该方法具有良好的寻优性能,因此基于Oracle罚函数方法的自适应约束差分进化算法是一种有效约束优化方法。     

基于HSV色彩空间的低照度图像增强

为解决低照度图像色彩偏暗、整体亮度较低、光照不均等问题,提出一种低照度图像增强算法.在HSV色彩空间将V分量按照灰度等级的高低进行分层处理,利用引入权重值的巴特沃斯滤波器对低灰度等级的V分量进行灰度级拉伸,利用提出的亮度控制方法对过度增强区域进行灰度级新映射,合并所有V分量并平滑,基于权重值和映射函数,提出S分量的自适应增强函数对图像色彩进行调整.实验结果表明,该算法是一种有效的低照度图像增强算法.     

基于降维扫描方法的自适应多目标遗传算法

为了有效地应用遗传算法解决H2/H∞鲁棒控制系统设计问题,将遗传算法与局部优化方法相结合,提出了基于降维扫描方法的自适应多目标遗传算法（DRSA-MOGA）。通过引入适应度函数标准化方法、基于最优Pareto解集搜索的降维扫描方法和适应度函数自适应调整方法,提高了算法的全局优化性能和局部搜索能力。仿真结果表明,DRSA-MOGA算法在不损失解的均匀度的情况下可以达到很高的逼近度。     

具有主从结构的粒子群优化算法

提出了一种具有主从结构的粒子群优化算法,该算法实现了惯性权重、加速因子、最大速度等系统参数与目标函数的同步优化。将主程序的一个粒子作为子程序的一组系统参数,在该组控制参数下使用基本的粒子群算法对子程序的目标函数进行优化,并把子程序优化所得的全局最优值返回主程序作为主程序的一个适应值,同时使用基本的粒子群算法对主程序的适应度函数进行优化。实验结果表明,该算法的优化性能较基本的粒子群算法有了显著提高。该方法对于粒子群算法的参数选择具有指导意义。     

求解函数优化问题的交谊舞算法

针对函数优化问题,借鉴交谊舞的有关机理,构造了若干舞蹈算子（空间移动,局部舞伴交换和概率舞池修正）.将优化过程视为在搜索空间中从候选解向最优解的信息交换过程,提出一种新的优化算法-交谊舞算法.着重研究了算法的原理机制,流程实现和改进方法.通过对复杂函数优化问题的仿真试验,结合与简单遗传算法对比分析,验证新算法具有有效防止陷入局部极小值,以及收敛速度快等性质.交谊舞算法具有解决函数优化问题的潜力.     

改进零参考深度曲线低照度图像增强算法

在低照度条件下拍摄的图像具有对比度低,亮度低,细节缺失等质量缺陷,给图像处理带来困难。提出一种改进零参考深度曲线低照度图像增强算法,通过在空间一致性损失函数中引入与卷积核大小相关参数,统一了不同尺寸图像的增强效果;将颜色不变损失、照明平滑损失函数与输入图像类型关联,使其增强效果的峰值信噪比提高17.75%,对比度提高26.75%;通过使用对称式卷积结构,解决原算法计算量大的问题;通过使用MobileNetV2轻量化网络对零参考深度网络(Zero-DCE)进行了优化,减少网络模型计算复杂度的同时保证模型较好的增强效果。     

Ray targeting for optimizing smooth freeform surfaces for LED non-rotational illumination

We propose an effective optimization method for generating smooth freeform surfaces for light-emitting diode(LED)non-rotational illumination based on ray targeting.This method begins with a starting design and goes through two optimization steps.An initial estimate is determined using a partial differential equation(PDE)method and a variable separation mapping.In the first optimization step the merit function is developed with ray targeting to ensure the shape of the illumination pattern.The purpose of the second optimization is to further improve the optical performance by constructing the merit function with uniformity and efficiency.Smooth freeform reflective and refractive surfaces,which can produce a uniform rectangular illumination without rotational symmetry,are designed using this method.The results show that uniform rectangular illumination is achieved and that smooth freeform surfaces are obtained.With ray targeting,the design efficiency can be significantly enhanced,and excellent optical performance can be achieved.     

Real-time robust background subtraction under rapidly changing illumination conditions

Fast robust background subtraction under sudden lighting changes is a challenging problem in many applications. In this paper, we propose a real-time approach, which combines the Eigenbackground and Statistical Illumination method to address this issue. The first algorithm is used to reconstruct the background frame, while the latter improves the foreground segmentation. In addition, we introduce an online spatial likelihood model by detecting reliable background pixels. Extensive quantitative experiments illustrate our approach consistently achieves significantly higher precision at high recall rates, compared to several state-of-the-art illumination invariant background subtraction methods.     

A method of illumination compensation for human face image based on quotient image

Illumination variation is one of the critical factors affecting face recognition rate. A novel approach for human face illumination compensation is presented in this paper. It constructs the nine-dimension face illumination subspace based on quotient image. In addition, with the aim to improve algorithm efficiency, a half-face illumination image is proposed and the low-dimension training set of the face image under different illumination conditions are obtained by means of PCA and wavelet transform. After processing, two different illumination compensation strategies are given: one is adding light, and the other is removing light. Based on the illumination compensation strategy, we implement the typical illumination sample image synthesis and the standard illumination sample image synthesis on a PCA feature subspace and a wavelet transform subspace, respectively, and the illumination compensation of the gray images and the color images are further realized. Experimental results based on the Yale Face Database B, the Extended Yale Face Database B and the CAS-PEAL Face Database indicate that execution time after compensation is approximately half the time and face recognition rate is improved by 20% compared with that of the original images.     

基于线性子空间和商图像理论的人脸光照补偿

光照是影响人脸识别率的主要因素,它已成为人脸识别技术发展的瓶颈。根据商图像理论,在所构造的低维训练集上对待识别的图像进行光照条件估计,通过加光和去光两种方法,实现了光照补偿的目的,并通过识别实验验证了其补偿效果。     

Integration of stamper fabrication and design optimization of LCD Light guides using silicon-based microfeatures

This study applies a novel fabrication process that combines anisotropic wet etching of silicon-on-insulator (SOI) wafers with electroforming to manufacture precision stampers. Micron-sized features, such as trapezoidal grooves and truncated pyramidal prisms, can be fabricated and distributed accurately. Because feature geometry and distribution can be accurately realized using the proposed scheme, design optimization of light guide plates (LGPs) becomes realistic. By observing the illumination characteristics of light emitting diode (LED) edge-lit LGPs, the distribution pattern of the LGP is transformed into a parameter design with seven anchor spacing and the spacing modulation amplitude of the micro features adjacent to LEDs. The proposed fuzzy optimization scheme manipulates distribution parameters to obtain an LGP design with high illumination uniformity. The design of a 3.5-inch LED edge-lit LGP is used as an illustrative example. The optical software program TracePro is applied to simulate luminance performance of BLM. The optimization converges rapidly and provides the optimum design with an average brightness of 2,266 (nit) and uniformity of 90% without use of diffusive sheets. Thus work demonstrates the feasibility and effectiveness of the proposed scheme.     

一种用于镜头检测的光照不变测度

光照问题是镜头检测中的一个重要的问题.光照变化严重干扰了许多镜头检测的精度.而现有的光照不变的帧间相似性测度计算代价高而降低了算法的速度.为此,提出了一种新的光照不变测度,并将它用于视频镜头检测.这种新的测度是基于图像中的边缘信息.它先提取了图像中的边缘,然后二值化,最后把背景的面积作为光照不变测度.理论分析及实验数据都证明了该测度的有效性.与同类测度相比,提高了速度而没有牺牲精度.     

Shadow compensation in 2D images for face recognition

Illumination variation that occurs on face images degrades the performance of face recognition. In this paper, we propose a novel approach to handling illumination variation for face recognition. Since most human faces are similar in shape, we can find the shadow characteristics, which the illumination variation makes on the faces depending on the direction of light. By using these characteristics, we can compensate for the illumination variation on face images. The proposed method is simple and requires much less computational effort than the other methods based on 3D models, and at the same time, provides a comparable recognition rate.     

支持向量机的多层动态自适应参数优化

首先提出了基于多层动态自适应搜索技术的最小二乘支持向量机参数优化方法，然后采用最小二乘支持向量机对典型非线性控制系统的辨识进行了研究．辨识结果表明，最小二乘支持向量机可以用于非线性控制系统辨识，多层动态自适应搜索方法确定了最优支持向量机参数，从而获得精确的非线性控制系统辨识结果．     

基于LS-SVM求解非线性常微分方程组的近似解

提出了一种基于最小二乘支持向量机(LS-SVM)的改进方法求解非线性常微分方程组初值问题的近似解.利用径向基核函数(RBF)可导的特点对LS-SVM模型进行改进,将含核函数导数形式的LS-SVM模型转化为优化问题进行求解.方法可在原始对偶集中获得近似解的最佳表示,所得近似解连续可微,且精度较高.给出数值算例,通过与真实解的对比验证了所提方法的准确性和有效性.     

短时交通流量预测的IPSO-BPNN算法

为了提高短时交通流量预测精度,利用相空间重构和预测模型参数间的相互联系,提出一种粒子群优化神经网络的短时交通流量预测模型。将相空间重构和预测模型参数编码为粒子群的粒子,短时交通流量预测精度作为粒子群的适应度函数,通过粒子之间协作获得预测模型全局最优参数,通过BP神经网络建立预测模型,利用短时交通流量数据对模型性能进行测试。结果表明,相对于传统参数优化方法,粒子群优化神经网络提高了短时交通流量的预测精度,为非线性预测问题提供了一种新的研究思路。