A fast converging sparse reconstruction algorithm in ghost imaging

A fast converging sparse reconstruction algorithm in ghost imaging is presented. It utilizes total variation regularization and its formulation is based on the Karush-Kuhn-Tucker (KKT) theorem in the theory of convex optimization. Tests using experimental data show that, compared with the algorithm of Gradient Projection for Sparse Reconstruction (GPSR), the proposed algorithm yields better results with less computation work.     

基于属性开运算的汽车牌照区域定位算法

现有的复杂背景下汽车牌照定位算法大都是基于牌照水平或者挖水平,处理倾斜情况下牌照定位是一个尚待解决的问题。文中提出了一种基于属性开运算的汽车牌照定位算法,其处理结果与牌照的倾斜角度无关,成功地应用于交通监探系统中。     

A fast Bayesian reconstruction algorithm for emission tomography with entropy prior converging to feasible images

The development and tests of an iterative reconstruction algorithm for emission tomography based on Bayesian statistical concepts are described. The algorithm uses the entropy of the generated image as a prior distribution, can be accelerated by the choice of an exponent, and converges uniformly to feasible images by the choice of one adjustable parameter. A feasible image has been defined as one that is consistent with the initial data (i.e. it is an image that, if truly a source of radiation in a patient, could have generated the initial data by the Poisson process that governs radioactive disintegration). The fundamental ideas of Bayesian reconstruction are discussed, along with the use of an entropy prior with an adjustable contrast parameter, the use of likelihood with data increment parameters as conditional probability, and the development of the new fast maximum a posteriori with entropy (FMAPE) Algorithm by the successive substitution method. It is shown that in the maximum likelihood estimator (MLE) and FMAPE algorithms, the only correct choice of initial image for the iterative procedure in the absence of a priori knowledge about the image configuration is a uniform field.     

A high-speed algorithm for elliptical object detection

A high-speed method for elliptical object location is proposed. Through the use of the geometric symmetry, it first locates the centers of elliptical objects and classifies the boundary points. Then, each point possible on an elliptical object boundary is used to obtain the remaining three parameters of the elliptical object. Since each boundary point is at most transformed to one point on the parameter space, the proposed method is very fast. Some experimental results are also given to show that the proposed method is better than some existing methods under the consideration of space and storage.     

一种实时多尺度目标检测识别算法

无人机目标检测与识别任务中,目标随着飞行高度的改变尺寸发生显著变化。常规目标检测模型中,获取的小目标细节信息有限,检测精度较低;而适用于小目标的实时检测模型往往容易丢失大目标的背景信息,降低大目标的检测精度。针对以上多尺度目标检测识别任务难点,提出一种基于改进特征金字塔网络(Feature Pyramid Network, FPN)结构的实时多尺度目标检测识别模型。该模型通过增加特征金字塔层级覆盖更广的目标尺度,获取更为丰富的目标信息;同时,利用跨连接增加不同尺度特征融合的多样性,降低特征传导距离,保留更加完整的尺度特征来提高模型检测识别多尺度目标的性能。通过实验发现,相比于原始网络结构和相同特征层级的四层特征金字塔结构,加入改进特征金字塔结构的多尺度目标检测模型识别性能得到了提升。     

A novel fast moving object contour tracking algorithm

If a somewhat fast moving object exists in a complicated tracking environment, snake’s nodes may fall into the inaccurate local minima. We propose a mean shift snake algorithm to solve this problem. However, if the object goes beyond the limits of mean shift snake module operation in suc- cessive sequences, mean shift snake’s nodes may also fall into the local minima in their moving to the new object position. This paper presents a motion compensation strategy by using particle filter; therefore a new Parti...     

一个快速有效的虹膜定位算法

传统的虹膜定位方法需要搜索的参数范围太大,速度较慢,并且,睫毛遮掩和外边缘模糊会大大影响定位效果,本文提出的定位算法基于从精到粗的思想,缩小参数范围,并有效的减少了这两个因素的影响.同现有两种常用方法通过试验比较,证明此算法能同时在精度和速度上有较大的改进.     

A rapidly converging algorithm for estimating respiratory mechanical parameters in a five-element model

A rapidly converging algorithm for computing values for respiratory mechanical parameters from forced random noise independance data was developed and verified. The algorithm, which was based on a five-element Mead-type model, minimized the sum of squared differences between the model's response and experimental data, while imposing a nonnegativity constraint on the parameter values. It yielded parameter values that showed excellent agreement with values obtained previously using standard nonlinear regression analysis, but required much less computer time, 10 s versus 1 h. When this algorithm is coupled with the forced random impedance data collection techniques, it provides a rapid noninvasive method for estimating respiratory inertance, central resistance, peripheral resistance, and airway compliance. The problem of estimating peripheral compliance was not solved by this algorithm.     

一种新的基于H.264压缩域的运动对象分割算法

提出了一种在H.264压缩域下进行运动对象分割的新算法。算法主要利用H.264码流中的运动矢量信息来进行对象分割,为了提高运动矢量信息的鲁棒性,首先利用I帧中的帧内预测模式和预测残差能量进行区域划分;在P帧中,利用帧间预测残差能量来更新区域划分结果,对部分区域的运动矢量进行归零化处理。再根据P帧中的分块模式,采用不同的滤波器对运动矢量进行滤波;最后利用滤波后的运动矢量信息建立对应的Gibbs势能函数,采用迭代条件模式方法求解最大后验概率,得到可靠的运动对象标记。实验结果表明,该运动对象分割算法可以获得有效并可靠的分割结果。     

基于Snake活动轮廓模型的视频跟踪分割方法

基于Snake活动轮廓模型,采用时空融合的方式,根据短时间内相邻帧的运动趋势差异相似的前提,首先将视频序列分成若干个小段,每段有k帧视频,选取段内的前两帧为关键帧,通过运动检测的方式自动得到这两帧中运动对象的大致区域;然后进行帧内Snake演变,搜索精确轮廓;最后以关键帧间运动对象形心的运动矢量预测勾勒后续帧的初始轮廓,进行帧内Snake精确轮廓定位,从而实现所有帧的视频对象分割。相比于传统方法,本文方法克服了手动绘制初始轮廓的缺点,在空域对Snake贪婪方法进行了改进而且精确度高,速度快。实验表明,本文方法成功地实现了前后帧图像之间运动对象的对应匹配关系,并通过改进后的Snake贪婪方法得到了精确的分割结果。     

一种超长级联国际传输专线故障机器定位算法的研究

国际电路不同于一般网内电路,具有跨多运营商、超长距离和级联组网的特性,这些特性给该类电路的故障定位带来挑战,但同时也形成其特有的规律。本文通过总结日常运维经验和分析各厂家设备告警信号传递规律,总结出可以用于计算机实现的国际电路故障定位算法。     

基于增强RPN的孪生网络目标跟踪算法

目前孪生网络跟踪器已经具有比较良好的表现,但是对于卷积神经网络所提取的特征仍没有较好地利用其特点,同时孪生网络通过相似性学习进行跟踪的特性使跟踪器的准确性和鲁棒性存在不足。提出了一种金字塔式特征融合的方法,根据骨干网络特征提取层不同深度具有不同侧重的特点提高网络对目标的表征能力,然后使用注意力机制对区域推荐网络（Region Proposal Network,RPN）进行增强,最终实现更精准更鲁棒的跟踪。在OTB100数据集的实验中,新提出的SiamERPN(Siamese Enhanced RPN)算法分别得到了0.668的成功率和0.876的精度,测试结果好于基线算法和其他对比算法。     

基于改进SSD的轻量化小目标检测算法

为提高SSD目标检测算法的小目标检测能力,提出在SSD算法中引入转置卷积结构,采用转置卷积将低分辨率高语义信息特征图与高分辨率低语义信息特征图相融合,增加低层特征提取能力,提高SSD算法的平均精准度。同时针对SSD算法存在模型过大,运行内存占用量过高,无法在嵌入式ARM设备上运行的问题,以DenseNet为基础,结合深度可分离卷积,逐点分组卷积与通道重排提出轻量化特征提取最小单元,将SSD算法特征提取部分替换为轻量化特征提取最小单元的组合后,可在嵌入式ARM设备上运行。在PASCAL VOC数据集和KITTI自动驾驶数据集上进行对比实验,结果表明改进后的网络结构在平均精准度上得到明显提升,模型参数数量得到有效降低。     

图像处理技术在火焰目标提取中的应用

在火灾识别系统中,为了精确地提取火焰目标,本文提出了一种综合有效的算法,该算法采用二维最大熵自动阈值法对火灾图像进行了分割处理,随后采用区域生长法,检测出火灾图像中的可疑区域,并以区域生长后的目标区域为模板,采用Otsu阈值选取法对当前图像进行精确分割,提取准确的火焰目标。实验证明该算法能够有效地提取出连续图像序列中的火焰目标图像。     

A hand tracking algorithm with particle filter and improved GVF snake model

To solve the problem that the accurate information of hand cannot be obtained by particle filter, a hand tracking algorithm based on particle filter combined with skin-color adaptive gradient vector flow (GVF) snake model is proposed. Adaptive GVF and skin color adaptive external guidance force are introduced to the traditional GVF snake model, guiding the curve to quickly converge to the deep concave region of hand contour and obtaining the complex hand contour accurately. This algorithm realizes a real-time correction of the particle filter parameters, avoiding the particle drift phenomenon. Experimental results show that the proposed algorithm can reduce the root mean square error of the hand tracking by 53%, and improve the accuracy of hand tracking in the case of complex and moving background, even with a large range of occlusion. This work has been supported by the National Natural Sciencal Foundation of China (No.61403274), and the Tianjin Technology Project of Intelligent Manufacturing (No.15ZXZNGX00160). E-mail:agwu@tju.edu.cn      

基于特征融合的粒子滤波目标跟踪新方法

针对传统粒子滤波(PF)算法采用单一颜色特征建模 跟踪目标性能差的缺陷,提出一种颜色特征与纹理特 征相融合的PF目标跟踪新算法。首先,采用一种具有抗噪声和保护纹理边缘的全局中值二值 模式 (GMBP)纹理算子,对模板图像进行局部差绝对值处理,得到幅 值序列模板,将幅值序列模板内的中值作为模板的阈值,与模板邻域比较获得新的纹理图像 ;然后,与 具有光照不变特性的局部二值模式(LBP)纹理算子结合,形成一种(GMLBP)新的纹理描述算子 。最后,分别计算GMLBP纹理特征粒子权值和HSV颜色特征粒子权 值,并依据权值大小确定融合系数,对纹理特征粒子权值和颜色特征粒子权值进行线 性融合,再对融合后粒子权值进行归一化处理,从而得到目标位置状态的最终估计值。对比 实验结果表明, 相对于单一颜色特征的目标跟踪算法,所提算法捕捉目标位置准确且具有更低的平均跟踪误 差,其平均误差降低了近2倍。     

基于深度学习的图像目标检测算法综述

图像目标检测是找出图像中感兴趣的目标，并确定他们的类别和位置，是当前计算机视觉领域的研究热点。近年来，由于深度学习在图像分类方面的准确度明显提高，基于深度学习的图像目标检测模型逐渐成为主流。首先介绍了图像目标检测模型中常用的卷积神经网络；然后，重点从候选区域、回归和anchor-free方法的角度对现有经典的图像目标检测模型进行综述；最后，根据在公共数据集上的检测结果分析模型的优势和缺点，总结了图像目标检测研究中存在的问题并对未来发展做出展望。     

基于时域定区间记忆补偿的视频对象分割算法

提出了一种新的基于时域定区间记忆补偿的视频对象分割算法。首先,使用对称帧帧差累计法及帧差图像的四阶矩检测出初始运动变化区域;然后,对检测出的初始运动变化区域通过时域定区间记忆补偿法进行补偿,并进一步整合形成全局运动记忆母板,在空域使用Sobel边缘检测算子较为精确地检测得到当前帧中所有边缘;最后,进行时空融合,从而提取出完整精细的运动对象轮廓并通过填充得到运动对象模板。实验证明了本文算法的正确性和快速性。     

A faster way to compute the noise-adjusted principal componentstransform matrix

The matrix for the noise-adjusted principal components (NAPC) transform is the solution of a generalized symmetric eigenvalue problem. Applied to remote sensing imagery, this entails the simultaneous diagonalization of data and noise covariance matrices. One of the two PC transforms of the original NAPC transform is replaced by several short, fast procedures. The total operation count for the computation of the NAPC transform matrix is halved     

A semi-automatic seeded region growing algorithm for video object localization and tracking

This paper describes a semi-automatic method for moving object segmentation and tracking. This method is suitable when a few objects have to be tracked, while the camera moves and fixates on them. The user delineates approximately the initial locations in a selected frame and specifies the depth ordering of the objects to be tracked. First, motion-based segmentation is obtained through an initial application of a region growing algorithm. The partition map is sequentially tracked from frame to frame using motion compensation and location prediction. The segmentation map is obtained by the region growing algorithm. Translational motion is assumed for the moving objects, and local intensity or color average may be used as additional features. A post-processing procedure regularizes the object boundaries over time.