Anchor-free的尺度自适应行人检测算法

Anchor作为行人检测算法中的初始框,可以解决行人平移问题和缓解行人尺度变化问题,目前的行人检测算法通常都基于anchor.然而,使用anchor就需要精心调整对检测性能影响非常大的anchor超参数,如anchor的尺度和高宽比等.为避免这一问题,提出一种基于anchor-free损失函数的行人检测算法,并通过融合...     

基于自适应状态转移的混合跟踪算法

为提高粒子滤波视觉目标跟踪算法的准确性和实时性,提出一种基于自适应状态转移的混合跟踪算法。首先采用零阶自适应变化模型来获取目标的可能状态,然后利用均值漂移算法的局部优化特性找到后验概率的最大值。在多峰值情况下由粒子滤波随机产生粒子,用新的粒子集来确定目标的最终位置。实验结果表明,这种改进的算法在保证准确性的同时,降低了系统的计算时间。     

Studying two-time scale processes in an adaptive system

Stability of an adaptive stabilization system with two-time scale processes is considered. The structure and properties of the fast motion loop are found. Features of the adaptive system with the filter for estimating derivatives of the output variables are determined. An example that illustrates the processes in the system is given.     

Tracking error bounds of adaptive nonstationary filtering

Adaptive filtering with error gradient algorithm and constant step-size is analyzed for a deterministic time variable optimum filtering vector. The unrealistic assumption of independent observations is replaced by a bounded memory model, largely justifiable in applications. Then the mean square tracking deviation (MSD) between the optimum vector and the algorithm output is proved to include two contributions; the stationary mode error, characteristic of convergence accuracy, which is proportional to the step-size; and the transient mode error, reflecting the rapidity of tracking, which is proportional to the squared ratio of the maximum optimum estimator increment to the step-size. This result agrees with the common intuition that there exists an optimum step-size which compromises between convergence accuracy and tracking speed.     

Tracking Context Changes through Meta-Learning

The article deals with the problem of learning incrementally (on-line) in domains where the target concepts are context-dependent, so that changes in context can produce more or less radical changes in the associated concepts. In particular, we concentrate on a class of learning tasks where the domain provides explicit clues as to the current context (e.g., attributes with characteristic values). A general two-level learning model is presented that effectively adjusts to changing contexts by trying to detect (via meta-learning) contextual clues and using this information to focus the learning process. Context learning and detection occur during regular on-line learning, without separate training phases for context recognition. Two operational systems based on this model are presented that differ in the underlying learning algorithm and in the way they use contextual information: METAL(B) combines meta-learning with a Bayesian classifier, while METAL(IB) is based on an instance-based learning algorithm. Experiments with synthetic domains as well as a number of real-world problems show that the algorithms are robust in a variety of dimensions, and that meta-learning can produce substantial increases in accuracy over simple object-level learning in situations with changing contexts.     

自适应理论在目标跟踪中的应用

在机动目标跟踪与定位中,结合EKF和自适应理论的优点和目标跟踪的非线性特征,提出了一种非线性系统的基于“当前”统计模型的自适应扩展卡尔曼滤波算法,根据机动目标的测量信息修正加速度方差,消除随机误差和噪声的干扰,提高预测的精度。通过Monte Carlo对比仿真实验表明该算法正确有效,定位精度较高,滤波效果得到改善,同时增强了稳定性,优于一般的EKF和MVEKF算法,为机动目标精确跟踪与定位的实现提供一种新的方法。     

Displacement estimates through adaptive affinities

A technique that is functionally equivalent to the oriented smoothing concept and reduces numerical complexity and computational costs by eliminating the smoothness requirement from the iteration process is introduced. Local affine transformations are applied to propagate uniquely computed flow vectors into homogeneous regions and along edges in a single step. The window within which the local affine transformation is performed can adapt to the local structure of the intensity pattern in accordance with the oriented smoothness concept as formulated by H. Nagel (1987)     

基于自适应粒子群优化算法的测试数据扩增方法

针对在回归测试中原有测试数据集往往难以满足新版本软件测试需求的问题,提出一种基于自适应粒子群算法（APSO）的测试数据扩增方法。首先,根据原有测试数据在新版本程序上的穿越路径与目标路径的相似度,在原有的测试数据集中选择合适的测试数据,作为初始种群的进化个体;然后,利用初始测试数据的穿越路径与目标路径的不同子路径,确定造成两者路径偏离的输入分量;最后,根据路径相似度构建适应度函数,利用APSO操作输入分量,生成新的测试数据。该方法针对四个基准程序与基于遗传算法（GA）和随机法的测试数据扩增方法相比,测试数据扩增效率分别平均提高了约56%和81%。实验结果表明,所提方法在回归测试方面有效地提高了测试数据扩增的效率,增强了其稳定性。     

Signal matching through scale space

Given a collection of similar signals that have been deformed with respect to each other, the general signal-matching problem is to recover the deformation. We formulate the problem as the minimization of an energy measure that combines a smoothness term and a similarity term. The minimization reduces to a dynamic system governed by a set of coupled, first-order differential equations. The dynamic system finds an optimal solution at a coarse scale and then tracks it continuously to a fine scale. Among the major themes in recent work on visual signal matching have been the notions of matching as constrained optimization, of variational surface reconstruction, and of coarse-to-fine matching. Our solution captures these in a precise, succinct, and unified form. Results are presented for one-dimensional signals, a motion sequence, and a stereo pair.     

多传感器跟踪系统自适应Kalman滤波融合

多传感器目标跟踪的一个实际问题是如何获得目标的过程噪声信息,以获得较好的跟踪性能。针对多传感器分布式估计融合系统,利用这种自适应技术给出了一种自适应Kalman滤波的融合方法,它具有与中心式相近的跟踪性能。计算机模拟结果表明：这种方法具有较优良的性能。     

近邻自适应局部尺度的谱聚类算法

提出一种在LUV空间中基于多层次化结构Nystrm方法的自适应谱聚类算法。首先引入LUV色彩空间,避免了RGB色彩空间中色彩辨别阈对分割的影响,在纹理、边缘区域取得了更好的分割效果;其次将谱聚类算法中基于多层次化结构的方法和基于Nystrm采样的方法结合起来,有效减少了运算时间、解决了数据量较大时计算过程中内存溢出的问题;最后在K均值聚类中通过对特征间隙(eigengap)的分析,自适应地选择K值的大小,解决了自动确定聚类数目的问题。将提出的方法在LUV色彩空间中和RGB色彩空间中分别进行图像分割实验,结果表明在LUV色彩空间中取得效果更加理想。同时也将提出的算法与基于Nystrm方法的谱聚类算法(spectral clustering-Nystrm,SC-N)进行比较。实验结果表明,该算法在数据运算量、运行时间和分割结果上都优于SC-N方法。     

Multiple Object Tracking through Background Learning

This paper discusses about the new approach of multiple object tracking relative to background information. The concept of multiple object tracking through background learning is based upon the theory of relativity, that involves a frame of reference in spatial domain to localize and/or track any object. The field of multiple object tracking has seen a lot of research, but researchers have considered the background as redundant. However, in object tracking, the background plays a vital role and leads to definite improvement in the overall process of tracking. In the present work an algorithm is proposed for the multiple object tracking through background learning. The learning framework is based on graph embedding approach for localizing multiple objects. The graph utilizes the inherent capabilities of depth modelling that assist in prior to track occlusion avoidance among multiple objects. The proposed algorithm has been compared with the recent work available in literature on numerous performance evaluation measures. It is observed that our proposed algorithm gives better performance.     

Tracking and parameter identification for model reference adaptive control

We provide barrier Lyapunov functions for model reference adaptive control algorithms, allowing us to prove robustness in the input‐to‐state stability framework and to compute rates of exponential convergence of the tracking and parameter identification errors to zero. Our results ensure identification of all entries of the unknown weight and control effectiveness matrices. We provide easily checked sufficient conditions for our relaxed persistency of excitation conditions to hold. Our illustrative numerical example demonstrates the performance of the control methods.     

Sliding mode control of singularly perturbed systems and its application in quad-rotors

This paper presents a sliding mode control scheme for tracking control of nonlinear singularly perturbed systems in the presence of model errors and external disturbances. A dual-loop feedback control is developed to provide accurate tracking capability and sufficient robustness to system uncertainties. A sliding mode controller is proposed in the outer-loop feedback design such that the plant states are stabilised for given reference trajectories, while an additional robust controller is designed in the inner loop to increase the adaptability to uncertainties, and reduce the effect of unmodelled high-frequency dynamics on plant dynamics. An appealing feature of the control scheme is the attenuation of chattering. The effectiveness and merits of the new control scheme developed are shown via a verification example of velocity control of a quad-rotor.     

A normal form augmentation approach to adaptive control of space robot systems

In this paper, we model a free-floating space robot system as anextended robot which is composed of a pseudo-arm representing the base motion resulting from six hyperthetic passive joints, and a real robot arm. The model allows us to categorize the space robot as an under-actuated system, and reveal fundamental properties of the system. Through input-output linearization of the model, we demonstrate a non-trivial internal dynamics, and propose an adaptive control scheme based on a normal form augmentation approach. This approach overcomes two fundamental difficulties in adaptive control design of space robot systems, i.e., nonlinear parameterization of the dynamic equation, and uncertainty of kinematic mapping from Cartesian space to joint space.     

Innately adaptive robotics through embodied evolution

Autonomous adaptation in robots has become recognised as crucial for devices deployed in remote or inhospitable environments. The aim of this work is to investigate autonomous robot adaptation, focussing on damage recovery and adaptation to unknown environments. An embodied evolutionary algorithm is introduced and its capabilities demonstrated with experimental results. This algorithm is shown to be able to control the motion of a robot snake effectively; this same algorithm inherently recovers the snake’s motion after damage. Another experiment shows that the algorithm is capable of contorting a shape-changing antenna in such a way as to minimise the affect of background noise on it, thus allowing the antenna to achieve a better signal.     

分离窗口快速尺度自适应目标跟踪算法

针对核相关滤波器（KCF）跟踪算法在面对尺度变化时产生的目标漂移问题，提出一种分离窗口快速尺度自适应目标跟踪算法——FSACF。首先，通过直接对原始帧图像进行特征提取得到基于显著性颜色特征的全局梯度组合特征图，以减小后续的尺度计算对性能的影响；其次，对全局特征图采用分离窗口法，自适应地选取尺度大小并计算对应的最大响应值；最后，采用定义的置信度函数自适应地更新迭代模板函数，提高模型的鲁棒性。通过带有不同干扰属性的视频集上进行实验，发现FSACF算法与KCF算法相比，在精度上提升7.4个百分点，成功率提高12.8个百分点；与未采用全局特征和分离窗口的算法对比，处理速度上提升1.5倍。实验结果表明，FSACF算法在尺度变化发生时能有效避免目标漂移的产生，同时具有一定的效率，并在精度与成功率上均优于对比算法。     

自适应尺度的局部强度聚类图像分割模型

针对传统活动轮廓模型无法精确分割强度不均匀图像，并且对尺度参数比较敏感的问题，提出了一种基于区域信息的自适应尺度的活动轮廓模型。根据图像的局部熵构建自适应尺度算子，利用图像的局部强度聚类性质构建能量函数。使用一组平滑基函数的线性组合来表示偏移场，这样可以增加模型的稳定性。通过最小化该能量，所提模型能够同时分割图像和估计偏移场，并且估计的偏移场可以用于强度不均匀校正。实验结果表明，与其它4种模型相比，该模型拥有更高的分割精确度，且分割结果对水平集函数的初始化和噪声具有鲁棒性。     

鲁棒的自适应尺度和方向的目标跟踪方法

针对在视频序列图像目标跟踪中,跟踪目标尺寸和跟踪目标相对背景运动的方位角都在实时变化,常规目标跟踪算法会引起尺度和方向定位偏差,导致跟踪漂移,甚至跟踪失败问题,提出鲁棒的目标尺度和方向自适应的跟踪方法。在Kalman滤波框架下,通过将运动目标的最小外接矩形信息转化为Kalman滤波参数,对目标运动进行建模。采用基于最小外接矩形的两步块匹配搜索方式实现对目标的中心定位,然后采用增量式搜索匹配方法根据最优尺度和角度的判别条件修正目标尺度和方向角度。通过动态评估不同目标模型在不同跟踪场景中的置信度,对目标模型进行动态更新。使用公用视频图像序列测试,实验结果验证了该方法的有效性。     

Tracking highly correlated targets through statistical multiplexing

We consider a special problem of multi-target tracking, where a group of targets are highly correlated, usually demonstrating a common motion pattern with individual variations. We focus on the task of searching and provide a statistical framework of embedding the correlation among targets and the most recent observations into sampling, where the correlation is learned dynamically from the previous tracking results. Proposal distribution is updated during the sampling process fused with the motion prior and observation information. In this way, the observation of a single target is multiplexed statistically through mutual correlation among the multiple targets, and the correlation serves as both a prior information to improve the efficiency and a constraint to prevent trackers from drifting. Extensive experiments on tracking both naturally correlated and environment-constrained targets demonstrate superior and promising robust results with low complexity.