The mean shift tracker has difficulty in tracking fast moving targets and suffers from tracking error accumulation problem. To overcome the limitations of the mean shift method, a new approach is proposed by integrati...The mean shift tracker has difficulty in tracking fast moving targets and suffers from tracking error accumulation problem. To overcome the limitations of the mean shift method, a new approach is proposed by integrating the mean shift algorithm and frame-difference methods. The rough position of the moving tar- get is first located by the direct frame-difference algorithm and three-frame-difference algorithm for the immobile camera scenes and mobile camera scenes, respectively. Then, the mean shift algorithm is used to achieve precise tracking of the target. Several tracking experiments show that the proposed method can effectively track first moving targets and overcome the tracking error accumulation problem.展开更多
In this paper,a compact difference scheme is established for the heat equations with multi-point boundary value conditions.The truncation error of the difference scheme is O(τ2+h^4),where t and h are the temporal ste...In this paper,a compact difference scheme is established for the heat equations with multi-point boundary value conditions.The truncation error of the difference scheme is O(τ2+h^4),where t and h are the temporal step size and the spatial step size.A prior estimate of the difference solution in a weighted norm is obtained.The unique solvability,stability and convergence of the difference scheme are proved by the energy method.The theoretical statements for the solution of the difference scheme are supported by numerical examples.展开更多
Kizmaz [13] studied the difference sequence spaces ∞（A）, c（A）, and co（A）. Several article dealt with the sets of sequences of m-th order difference of which are bounded, convergent, or convergent to zero. Alta...Kizmaz [13] studied the difference sequence spaces ∞（A）, c（A）, and co（A）. Several article dealt with the sets of sequences of m-th order difference of which are bounded, convergent, or convergent to zero. Altay and Basar [5] and Altay, Basar, and Mursaleen [7] introduced the Euler sequence spaces e0^r, ec^r, and e∞^r, respectively. The main purpose of this article is to introduce the spaces e0^r△^（m））, ec^r△^（m））, and e∞^r△^（m））consisting of all sequences whose mth order differences are in the Euler spaces e0^r, ec^r, and e∞^r, respectively. Moreover, the authors give some topological properties and inclusion relations, and determine the α-, β-, and γ-duals of the spaces e0^r△^（m））, ec^r△^（m））, and e∞^r△^（m））, and the Schauder basis of the spaces e0^r△^（m））, ec^r△^（m））. The last section of the article is devoted to the characterization of some matrix mappings on the sequence space ec^r△^（m））.展开更多
Key frame extraction based on sparse coding can reduce the redundancy of continuous frames and concisely express the entire video.However,how to develop a key frame extraction algorithm that can automatically extract ...Key frame extraction based on sparse coding can reduce the redundancy of continuous frames and concisely express the entire video.However,how to develop a key frame extraction algorithm that can automatically extract a few frames with a low reconstruction error remains a challenge.In this paper,we propose a novel model of structured sparse-codingbased key frame extraction,wherein a nonconvex group log-regularizer is used with strong sparsity and a low reconstruction error.To automatically extract key frames,a decomposition scheme is designed to separate the sparse coefficient matrix by rows.The rows enforced by the nonconvex group log-regularizer become zero or nonzero,leading to the learning of the structured sparse coefficient matrix.To solve the nonconvex problems due to the log-regularizer,the difference of convex algorithm(DCA)is employed to decompose the log-regularizer into the difference of two convex functions related to the l1 norm,which can be directly obtained through the proximal operator.Therefore,an efficient structured sparse coding algorithm with the group log-regularizer for key frame extraction is developed,which can automatically extract a few frames directly from the video to represent the entire video with a low reconstruction error.Experimental results demonstrate that the proposed algorithm can extract more accurate key frames from most Sum Me videos compared to the stateof-the-art methods.Furthermore,the proposed algorithm can obtain a higher compression with a nearly 18% increase compared to sparse modeling representation selection(SMRS)and an 8% increase compared to SC-det on the VSUMM dataset.展开更多
In this article, using generalized weighted mean and difference matrix of order m, we introduce the paranormed sequence space l（u, v, p; △（m））, which consist of the sequences whose generalized weighted △（m）-di...In this article, using generalized weighted mean and difference matrix of order m, we introduce the paranormed sequence space l（u, v, p; △（m））, which consist of the sequences whose generalized weighted △（m）-difference means are in the linear space l（p） defined by I.J.Maddox. Also, we determine the basis of this space and compute its α-, β- and γ-duals. Further, we give the characterization of the classes of matrix mappings from l（u, v, p, △（m）） to l∞, c, and co. Finally, we apply the Hausdorff measure of noncompacness to characterize some classes of compact operators given by matrices on the space lp（U, v, △（m）） （1 ≤ p 〈 ∞）.展开更多
Men and women applied language distinct from each other in many ways. The thesis gives an illustration of gender differences in conversation and different interpretive frames within which the discourse between men and...Men and women applied language distinct from each other in many ways. The thesis gives an illustration of gender differences in conversation and different interpretive frames within which the discourse between men and women take place. More profoundly,it tries to explain them from perspective of socialization.展开更多
针对人体目标检测中目标处于图像边缘或目标较小时,MobileNet-SSD算法检测效果不佳的问题,提出一种基于MobileNet-SSD算法与帧差法结合的人体目标检测方法。首先,基于帧差法实现运动目标粗略位置的获取;其次以粗略位置为基准计算目标运...针对人体目标检测中目标处于图像边缘或目标较小时,MobileNet-SSD算法检测效果不佳的问题,提出一种基于MobileNet-SSD算法与帧差法结合的人体目标检测方法。首先,基于帧差法实现运动目标粗略位置的获取;其次以粗略位置为基准计算目标运动区域,并在检测原图中截取该区域,实现自适应感兴趣区域(Region of Interesting,ROI)的获取,并将ROI送入MobileNet-SSD模型中实现人体目标检测;最后使用MobileNet-SSD算法与帧差法结合的人体目标检测方法开展人体目标检测试验。结果表明:基于MobileNet-SSD与帧差法结合的人体目标检测方法能有效地检测出图像边缘区域的人体目标,且不影响原有检测速度。展开更多
针对在实际场景中,人体与背景之间存在相似的颜色和纹理,且人体运动涉及姿态的多样性,在其复杂多变的背景下运动,使分割出人体轨迹较为困难的问题,提出对称差分算法下人体运动轨迹图像分割技术。采用七帧对称差分算法提取人体运动图像...针对在实际场景中,人体与背景之间存在相似的颜色和纹理,且人体运动涉及姿态的多样性,在其复杂多变的背景下运动,使分割出人体轨迹较为困难的问题,提出对称差分算法下人体运动轨迹图像分割技术。采用七帧对称差分算法提取人体运动图像序列的前3帧和后3帧图像,计算其绝对差分图像,获取人体运动目标区域;采用非参数的统计迭代(Mean Shift)算法提取像素模值点分布情况,生成超像素,利用非参数贝叶斯聚类模型融合超像素提取人体运动目标轮廓;利用高斯混合模型建立人体运动轨迹模型,采用极限学习机求解模型识别人体运动轨迹,实现人体运动轨迹图像分割。实验结果表明,所提方法IOU(Intersection Over Union)值最高可达97%,提取运动目标区域和识别运动轨迹精度较高、分割性能较好,适用于人体运动轨迹图像分割。展开更多
基金supported by the Fundamental Research Funds for the Central Universities Project(CDJZR10170010)
文摘The mean shift tracker has difficulty in tracking fast moving targets and suffers from tracking error accumulation problem. To overcome the limitations of the mean shift method, a new approach is proposed by integrating the mean shift algorithm and frame-difference methods. The rough position of the moving tar- get is first located by the direct frame-difference algorithm and three-frame-difference algorithm for the immobile camera scenes and mobile camera scenes, respectively. Then, the mean shift algorithm is used to achieve precise tracking of the target. Several tracking experiments show that the proposed method can effectively track first moving targets and overcome the tracking error accumulation problem.
基金The research is supported by the National Natural Science Foundation of China(No.11671081)the Fundamental Research Funds for the Central Universities(No.242017K41044).
文摘In this paper,a compact difference scheme is established for the heat equations with multi-point boundary value conditions.The truncation error of the difference scheme is O(τ2+h^4),where t and h are the temporal step size and the spatial step size.A prior estimate of the difference solution in a weighted norm is obtained.The unique solvability,stability and convergence of the difference scheme are proved by the energy method.The theoretical statements for the solution of the difference scheme are supported by numerical examples.
文摘Kizmaz [13] studied the difference sequence spaces ∞（A）, c（A）, and co（A）. Several article dealt with the sets of sequences of m-th order difference of which are bounded, convergent, or convergent to zero. Altay and Basar [5] and Altay, Basar, and Mursaleen [7] introduced the Euler sequence spaces e0^r, ec^r, and e∞^r, respectively. The main purpose of this article is to introduce the spaces e0^r△^（m））, ec^r△^（m））, and e∞^r△^（m））consisting of all sequences whose mth order differences are in the Euler spaces e0^r, ec^r, and e∞^r, respectively. Moreover, the authors give some topological properties and inclusion relations, and determine the α-, β-, and γ-duals of the spaces e0^r△^（m））, ec^r△^（m））, and e∞^r△^（m））, and the Schauder basis of the spaces e0^r△^（m））, ec^r△^（m））. The last section of the article is devoted to the characterization of some matrix mappings on the sequence space ec^r△^（m））.
基金supported in part by the National Natural Science Foundation of China(61903090,61727810,62073086,62076077,61803096,U191140003)the Guangzhou Science and Technology Program Project(202002030289)Japan Society for the Promotion of Science(JSPS)KAKENHI(18K18083)。
文摘Key frame extraction based on sparse coding can reduce the redundancy of continuous frames and concisely express the entire video.However,how to develop a key frame extraction algorithm that can automatically extract a few frames with a low reconstruction error remains a challenge.In this paper,we propose a novel model of structured sparse-codingbased key frame extraction,wherein a nonconvex group log-regularizer is used with strong sparsity and a low reconstruction error.To automatically extract key frames,a decomposition scheme is designed to separate the sparse coefficient matrix by rows.The rows enforced by the nonconvex group log-regularizer become zero or nonzero,leading to the learning of the structured sparse coefficient matrix.To solve the nonconvex problems due to the log-regularizer,the difference of convex algorithm(DCA)is employed to decompose the log-regularizer into the difference of two convex functions related to the l1 norm,which can be directly obtained through the proximal operator.Therefore,an efficient structured sparse coding algorithm with the group log-regularizer for key frame extraction is developed,which can automatically extract a few frames directly from the video to represent the entire video with a low reconstruction error.Experimental results demonstrate that the proposed algorithm can extract more accurate key frames from most Sum Me videos compared to the stateof-the-art methods.Furthermore,the proposed algorithm can obtain a higher compression with a nearly 18% increase compared to sparse modeling representation selection(SMRS)and an 8% increase compared to SC-det on the VSUMM dataset.
文摘In this article, using generalized weighted mean and difference matrix of order m, we introduce the paranormed sequence space l（u, v, p; △（m））, which consist of the sequences whose generalized weighted △（m）-difference means are in the linear space l（p） defined by I.J.Maddox. Also, we determine the basis of this space and compute its α-, β- and γ-duals. Further, we give the characterization of the classes of matrix mappings from l（u, v, p, △（m）） to l∞, c, and co. Finally, we apply the Hausdorff measure of noncompacness to characterize some classes of compact operators given by matrices on the space lp（U, v, △（m）） （1 ≤ p 〈 ∞）.
文摘Men and women applied language distinct from each other in many ways. The thesis gives an illustration of gender differences in conversation and different interpretive frames within which the discourse between men and women take place. More profoundly,it tries to explain them from perspective of socialization.
文摘针对人体目标检测中目标处于图像边缘或目标较小时,MobileNet-SSD算法检测效果不佳的问题,提出一种基于MobileNet-SSD算法与帧差法结合的人体目标检测方法。首先,基于帧差法实现运动目标粗略位置的获取;其次以粗略位置为基准计算目标运动区域,并在检测原图中截取该区域,实现自适应感兴趣区域(Region of Interesting,ROI)的获取,并将ROI送入MobileNet-SSD模型中实现人体目标检测;最后使用MobileNet-SSD算法与帧差法结合的人体目标检测方法开展人体目标检测试验。结果表明:基于MobileNet-SSD与帧差法结合的人体目标检测方法能有效地检测出图像边缘区域的人体目标,且不影响原有检测速度。
文摘针对在实际场景中,人体与背景之间存在相似的颜色和纹理,且人体运动涉及姿态的多样性,在其复杂多变的背景下运动,使分割出人体轨迹较为困难的问题,提出对称差分算法下人体运动轨迹图像分割技术。采用七帧对称差分算法提取人体运动图像序列的前3帧和后3帧图像,计算其绝对差分图像,获取人体运动目标区域;采用非参数的统计迭代(Mean Shift)算法提取像素模值点分布情况,生成超像素,利用非参数贝叶斯聚类模型融合超像素提取人体运动目标轮廓;利用高斯混合模型建立人体运动轨迹模型,采用极限学习机求解模型识别人体运动轨迹,实现人体运动轨迹图像分割。实验结果表明,所提方法IOU(Intersection Over Union)值最高可达97%,提取运动目标区域和识别运动轨迹精度较高、分割性能较好,适用于人体运动轨迹图像分割。
