基于视网膜大细胞通路计算模型和深度学习的无人机目标检测

在复杂背景下的小型无人机红外目标检测是计算机视觉领域的挑战性课题.传统目标检测算法利用深度卷积神经网络提取无人机的静态外观特征并进行模式判别,但在复杂背景下且目标外观不清晰时的性能会显著下降.本文借鉴生物视网膜机制,通过视网膜大细胞通路模型提取无人机目标的时空运动信息,同时借助深度卷积神经网络获得基于静态表观特征的目标置信度图,进而将视网膜时空运动信息与深度卷积网络的目标置信度图进行融合获得目标检测结果.在Anti-UAV2020公开数据集上的评估结果表明,所提出算法的检测精确率达到86.90%,超过了业内主流的YOLO-v3算法.     

基于监控视频的前景目标提取

对含有动、静态背景的稳定图像处理时,对比了主成分追踪鲁棒主成分分析法(RPCA)、贝叶斯鲁棒主成分分析法(Bayesian RPCA)和高斯混合模型的鲁棒主成分分析法(MoG-RPCA),3种方法对静态背景下的前景提取都较为完整.而动态背景下只有Bayesian RPCA和MoG-RPCA提取出了完整的前景目标,但是Bayesian RPCA计算速度很慢,且不能够处理复杂噪声.所以MoG-RPCA模型更具有对复杂噪声的适应性,动、静态背景情况下均提取出精度较高的前景目标,且具有较快的计算速度.当图像不稳定时,采用改进的MoG-RPCA模型对非稳定拍摄的抖动视频进行前景目标提取,并在第197帧抖动图像中清晰地提取出显著前景目标,且运算速度较快.在为了快速找到目标出现的帧时,对高斯混合模型背景差分法进行改进,利用K-means聚类算法快速得到聚类中心点,然后作为高斯混合模型背景更新时的初始化均值参数,从而提高在复杂场景下前景目标的检测精度.对于多角度追踪任务,不同角度、近似同一地点的多个监控视频图像中前景目标的提取,可采用跨摄像头视角跟踪结果融合的方法,然后对目标进行匹配.     

多模型加权融合机制的石墨纯度识别

针对人工识别的效率低及单一卷积神经网络提取特征的遗漏问题,提出了多模型加权融合机制的石墨纯度识别算法.在自建小样本数据集上,进行离线扩充和在线增强,提高模型的泛化能力,减少深层CNN的过拟合问题;结合迁移学习,利用优化的AlexNet和ResNet50构建双通道卷积神经网络,提取石墨图像的深层次特征,并将两者的特征进行...     

基于Kriging模型和提升小波变换的随机模型修正北大核心CSCD

为提高随机模型修正效率,减小计算量,提出了一种基于Kriging模型和提升小波变换的随机模型修正方法.首先,对加速度频响函数进行提升小波变换,提取第5层近似系数代替原频响函数.其次,采用拉丁超立方抽样抽取待修正样本,将其作为Kriging模型的输入,对应的近似系数作为输出,构建Kriging模型.提出了一种引入莱维飞行(Lévy flight)的蝴蝶优化算法(LBOA),并将其应用于Kriging模型相关参数的寻优中,提高Kriging模型的精度.最后,以最小化Wasserstein距离为目标,通过鲸鱼优化算法求解待修正参数的均值.测试函数结果表明,LBOA在寻优能力、收敛精度和稳定性等方面有了很大的提升.数值算例的修正误差均低于0.4%,验证了所提模型修正方法具有较高的修正精度和效率.     

基于smoothL1改进的边框回归损失函数

为了进一步提高目标检测任务中的边框回归精度,提出一种基于smoothL1改进的边框回归损失函数.通过自适应地增大smoothL1中非离散点的梯度,缓解了模型反向传播中对离散点和非离散点梯度分布不平衡问题,提高了模型的精度.实验结果表明,在PASCAL VOC2007测试集上,基于改进的smoothL1的目标检测模型Faster R-CNN,平均精度均值(mAP)达到了70.8％,相较smoothL1,模型精度有所提高.     

电力巡检中绝缘子缺陷问题研究

针对电力巡检中由于无人机航拍绝缘子图像尺寸大,导致小目标检测难的问题,提出一种结合滑动窗口和主流神经网络的改进算法,对大尺寸图像中绝缘子串珠的分割以及缺陷区域的定位进行研究.在绝缘子串珠分割问题上,采用滑动窗口结合Mask R-CNN神经网络算法进行图像分割;在绝缘子缺陷检测问题上,结合滑动窗口和非极大值抑制算法对Yolo-v4神经网络进行改进,提高绝缘子缺陷区域小目标检测的精度.实验结果表明,分割所得绝缘子串珠掩膜图与标准掩膜图的相似度(Dice系数)为85.68%,与人工标记的绝缘子区域相比达到更高的相似度,绝缘子缺陷定位的平均精度均值(mean AP)为94.88%,与真实数据相比,平均重叠度为86.93%.     

基于LBP算子和类覆盖捕获图的人脸检测算法

针对传统人脸检测中的过分类问题,提出一种结合LBP算子与类覆盖捕获图的人脸检测算法.该算法首先用ε-LBP算子提取人脸图像纹理特征,并把对应不同ε值提取的LBP特征数据加权融合起来,形成人脸图像特征向量,然后采用类覆盖捕获图构造分类器,最终对人脸图像实现有效检测.与传统方法相比,基于随机图理论的类覆盖捕获图能够克服过分类缺陷,比其他近邻图分类器更具优势,性能也比较稳定.实验结果表明,该算法可以有效检测人脸图像,尤其对存在模糊和光照异常的人脸图像具有较高的精确度和鲁棒性.     

基于特征选择和梯度提升算法的高光谱遥感地物识别

高光谱遥感数据波段数目较多,且波段之间的相关性高,影响到敏感波段在地物识别中的作用,并造成大量冗余计算,降低时效.提出了一种随机森林结合递归特征消除的敏感特征选择方案,以提高高光谱遥感地物识别的精度与效率.通过RF-RFE特征选择方法得到最优特征组合,并运用LightGBM和XGBoost等提升算法来提高分类精度.在江苏省常州的茶树数据集上进行分类实验时,在原始数据上的分类精度达到了94.27%和94.45%;在特征选择出的最优特征子集上进行实验时,分类精度达到了94.40%和94.36%.实验结果表明,该方案的分类精度要优于决策树和朴素贝叶斯等传统分类算法,同时大幅减少了运算量,取得了较好的识别效果,具有一定的推广和应用价值.     

一种改进灰狼优化算法研究及应用

针对灰狼算法易陷入局部最优、收敛精度不高、收敛速度慢等缺点,提出一种改进的灰狼算法.引入莱维飞行,扩大搜索范围,增强全局搜索能力,避免陷入局部最优;引入贪婪原理,提升种群优良性以提高算法收敛精度;引入自适应收敛因子,加快收敛速度;引入动态权重策略,制约全局搜索与局部搜索的相互影响.将改进算法与其他四种算法作对比,实验表明,改进算法在收敛速度与收敛精度上都有更好的性能.最后,应用于图像多阈值分割中,采用GWO-Otsu法可以克服传统Otsu法在多阈值分割时计算量大,实时性差的特点,不但能够取得最优解,且明显缩减计算时间.     

基于多尺度局部结构约束的特征点匹配算法

对于含有重复模式较多、低重叠区域的图像,特征匹配时会存在离群点较多及伪同构等问题,从而影响匹配的精度.为了提高匹配算法的精度和鲁棒性,本文提出了一种基于多尺度局部结构相似性的特征点匹配算法(MLSS).首先定义局部结构相似性度量方法,即多尺度近邻结构相似性及多尺度局部拓扑结构相似性.基于局部结构相似性,构造特征匹配模型,并采用确定性退火思想去除离群点,实现特征点集的精确和快速的匹配.使用Daisy数据集、VGG数据集和航空遥感数据集中的22组典型数据对算法的性能进行验证,匹配结果显示,该算法的精度和鲁棒性优于其他四种经典的特征匹配方法.     

Hom-Hopf代数上的余代数结构

Let(C, α) and(H, β) be Hom-bialgebras and ω : C H → H  C a linear map.We introduce the concept of a Hom-ω-crossed coproduct(Cω σ H, γ) and we give necessary and sufficient conditions for the new object to be a Hom-Hopf algebra.     

Tractability of Approximation for Weighted Korobov Spaces on Classical and Quantum Computers

We study the approximation problem (or problem of optimal recovery in the $L_2$-norm) for weighted Korobov spaces with smoothness parameter $\a$. The weights $\gamma_j$ of the Korobov spaces moderate the behavior of periodic functions with respect to successive variables. The nonnegative smoothness parameter $\a$ measures the decay of Fourier coefficients. For $\a=0$, the Korobov space is the $L_2$ space, whereas for positive $\a$, the Korobov space is a space of periodic functions with some smoothness and the approximation problem corresponds to a compact operator. The periodic functions are defined on $[0,1]^d$ and our main interest is when the dimension $d$ varies and may be large. We consider algorithms using two different classes of information. The first class $\lall$ consists of arbitrary linear functionals. The second class $\lstd$ consists of only function values and this class is more realistic in practical computations. We want to know when the approximation problem is tractable. Tractability means that there exists an algorithm whose error is at most $\e$ and whose information cost is bounded by a polynomial in the dimension $d$ and in $\e^{-1}$. Strong tractability means that the bound does not depend on $d$ and is polynomial in $\e^{-1}$. In this paper we consider the worst case, randomized, and quantum settings. In each setting, the concepts of error and cost are defined differently and, therefore, tractability and strong tractability depend on the setting and on the class of information. In the worst case setting, we apply known results to prove that strong tractability and tractability in the class $\lall$ are equivalent. This holds if and only if $\a>0$ and the sum-exponent $s_{\g}$ of weights is finite, where $s_{\g}= \inf\{s>0 : \xxsum_{j=1}^\infty\g_j^s\,<\,\infty\}$. In the worst case setting for the class $\lstd$ we must assume that $\a>1$ to guarantee that functionals from $\lstd$ are continuous. The notions of strong tractability and tractability are not equivalent. In particular, strong tractability holds if and only if $\a>1$ and $\xxsum_{j=1}^\infty\g_j<\infty$. In the randomized setting, it is known that randomization does not help over the worst case setting in the class $\lall$. For the class $\lstd$, we prove that strong tractability and tractability are equivalent and this holds under the same assumption as for the class $\lall$ in the worst case setting, that is, if and only if $\a>0$ and $s_{\g} < \infty$. In the quantum setting, we consider only upper bounds for the class $\lstd$ with $\a>1$. We prove that $s_{\g}<\infty$ implies strong tractability. Hence for $s_{\g}>1$, the randomized and quantum settings both break worst case intractability of approximation for the class $\lstd$. We indicate cost bounds on algorithms with error at most $\e$. Let $\cc(d)$ denote the cost of computing $L(f)$ for $L\in \lall$ or $L\in \lstd$, and let the cost of one arithmetic operation be taken as unity. The information cost bound in the worst case setting for the class $\lall$ is of order $\cc (d) \cdot \e^{-p}$ with $p$ being roughly equal to $2\max(s_\g,\a^{-1})$. Then for the class $\lstd$ in the randomized setting, we present an algorithm with error at most $\e$ and whose total cost is of order $\cc(d)\e^{-p-2} + d\e^{-2p-2}$, which for small $\e$ is roughly $$ d\e^{-2p-2}. $$ In the quantum setting, we present a quantum algorithm with error at most $\e$ that uses about only $d + \log \e^{-1}$ qubits and whose total cost is of order $$ (\cc(d) +d) \e^{-1-3p/2}. $$ The ratio of the costs of the algorithms in the quantum setting and the randomized setting is of order $$ \frac{d}{\cc(d)+d}\,\left(\frac1{\e}\right)^{1+p/2}. $$ Hence, we have a polynomial speedup of order $\e^{-(1+p/2)}$. We stress that $p$ can be arbitrarily large, and in this case the speedup is huge.     

一般型三维簇的三典范映射

研究了一般型三维射影代数簇的三典范映射性质.当几何亏格p_g≥2时,整数m≥4的多典范映射结构已经比较清晰,但是三典范映射的结构还有很多未知问题.如果三典范映射φ_3不是双有理的,本文证明了φ_3的映射次数(如果φ_3是一般有限映射)或者一般纤维中一般不可约分支的几何亏格(如果φ_3是纤维型)有一个上界.     

Intermittent Behaviors in Coupled Piecewise Expanding Map Lattices

In this paper, we propose a new method to study intermittent behaviors of coupled piecewise-expanding map lattices. We show that the successive transition between ordered and disordered phases occurs for almost every orbit when the coupling is small. That is,where $x_i(n)$ correspond to the coordinates of $m$ nodes at the iterative step $n$. Moreover, when the uncoupled system is generated by the tent map and the lattice consists of two nodes, we prove a phase transition occurs between synchronization and intermittent behaviors. That is, $$\lim_{n\rightarrow \infty}| x_1(n)-x_2(n)|=0\quad\text{for }\ \ \Big|c-\frac12\Big|<\frac14$$ and intermittent behaviors occur for $|c-\frac12|>\frac14$, where $0\le c\le 1$ is the coupling.     

(Semi-)Nonrelativisitic Limit of the Nonlinear Dirac Equations

We consider the nonlinear Dirac equation (NLD) with time dependent external electro-magnetic potentials, involving a dimensionless parameter $ε\in(0,1]$ which is inversely proportional to the speed of light. In the nonrelativistic limit regime $ε\ll1$ (speed of light tends to infinity), we decompose the solution into the eigenspaces associated with the 'free Dirac operator' and construct an approximation to the NLD with $O(ε^2)$ error. The NLD converges (with a phase factor) to a coupled nonlinear Schrödinger system (NLS) with external electric potential in the nonrelativistic limit as $ε\to0^+$, and the error of the NLS approximation is first order $O(ε)$. The constructed $O(ε^2)$ approximation is well-suited for numerical purposes.     

严格上三角矩阵李代数上的积零导子

设$F$ 为域, $n\geq 3$, $\bf{N}$$(n,\mathbb{F})$ 为域$\mathbb{F}$ 上所有$n\times n$ 阶严格上三角矩阵构成的严格上三角矩阵李代数, 其李运算为$[x,y]=xy-yx$. $\bf{N}$$(n, \mathbb{F})$ 上一线性映射$\varphi$ 称为积零导子,如果由$[x,y]=0, x,y\in \bf{N}$$(n,\mathbb{F})$,总可推出 $[\varphi(x), y]+[x,\varphi(y)]=0$. 本文证明 $\bf{N}$$(n,\mathbb{F})$上一线性映射 $\varphi$ 为积零导子当且仅当 $\varphi$ 为$\bf{N}$$(n,\mathbb{F})$ 上内导子, 对角线导子, 极端导子, 中心导子和标量乘法的和.     

基于DC规划方法的稀疏临近支持向量机

为了提高临近支持向量机（PSVM）的数值表现,在PSVM的模型中引入了$\ell_0$-范数正则项,提出了稀疏临近支持向量机模型（SPSVM）,从而提高分类器的特征选择能力。然而带有$\ell_0$-范数正则项的问题往往是NP-难问题,为了克服这一问题,采用非凸连续函数近似$\ell_0$-范数,并通过适当的DC分解将问题转化成DC规划问题进行求解,同时还讨论了算法的收敛性。数值实验结果表明不论是在仿真数据还是在实际数据中,所提出的方法是比较有效稳定的。     

全矩阵代数的抛物子代数上具有可导性的非线性映射

Let F be a field of characteristic 0, Mn（F） the full matrix algebra over F, t the subalgebra of Mn（F） consisting of all upper triangular matrices. Any subalgebra of Mn（F） containing t is called a parabolic subalgebra of Mn（F）. Let P be a parabolic subalgebra of Mn（F）. A map φ on P is said to satisfy derivability if φ（x·y） = φ（x）·y＋x·φ（y） for all x,y ∈ P, where φ is not necessarily linear. Note that a map satisfying derivability on P is not necessarily a derivation on P. In this paper, we prove that a map φ on P satisfies derivability if and only if φ is a sum of an inner derivation and an additive quasi-derivation on P. In particular, any derivation of parabolic subalgebras of Mn（F） is an inner derivation.     

Poincar{\'e} Bifurcation from an Elliptic Hamiltonian of Degree Four with Two-saddle Cycle

In this paper, we consider Poincar{\''e} bifurcation from an elliptic Hamiltonian of degree four with two-saddle cycle. Based on the Chebyshev criterion, not only one case in the Li{\''e}nard equations of type $(3, 2)$ is discussed again in a different way from the previous ones, but also its two extended cases are investigated, where the perturbations are given respectively by adding $\varepsilon y(d_0 + d_2 v^{2n})\frac{\partial }{{\partial y}}$ with $ n\in \mathbb{N^+}$ and $\varepsilon y(d_0 + d_4 {v^4}+ d_2 v^{2n+4})\frac{\partial }{{\partial y}}$ with $n=-1$ or $ n\in \mathbb{N^+}$, for small $\varepsilon > 0$. For the above cases, we obtain all the sharp upper bound of the number of zeros for Abelian integrals, from which the existence of limit cycles at most via the first-order Melnikov functions is determined. Finally, one example of double limit cycles for the latter case is given.     

标准算子代数上保因子交换性的映射

Let X, Y be real or complex Banach spaces with dimension greater than 2 and A, B be standard operator algebras on X and Y, respectively. Let φ ：A →B be a unital surjective map. In this paper, we characterize the map φ on .4 which satisfies （A - B）R = R（A-B） ξR （（A-B）→ （φ（B））φ（R） =φ（R）（（A）- （B）） for A, B, R E .4 and for some scalar