We propose a novel online multiple object tracker taking structure information into account. State-of-the-art multi-object tracking (MOT) approaches commonly focus on discriminative appearance features, while neglect in different levels structure information and the core of data association. Addressing this, we design a new tracker fully exploiting structure information and encoding such information into the cost function of the graph matching model. Firstly, a new measurement is proposed to compare the structure similarity of two graphs whose nodes are equal. With this measurement, we define a complete matching which performs association in high efficiency. Secondly, for incomplete matching scenarios, a structure keeper net (SKnet) is designed to adaptively establish the graph for matching. Finally, we conduct extensive experiments on benchmarks including MOT2015 and MOT17. The results demonstrate the competitiveness and practicability of our tracker.
In this article, the kernel-based methods explained by a graph embedding framework are analyzed and their nature is revealed,
i.e. any kernel-based method in a graph embedding framework is equivalent to kernel principal component analysis plus its
corresponding linear one. Based on this result, the authors propose a complete kernel-based algorithms framework. Any algorithm
in our framework makes full use of two kinds of discriminant information, irregular and regular. The proposed algorithms framework
is tested and evaluated using the ORL, Yale and FERET face databases. The experiment results demonstrate the effectiveness
of our proposed algorithms framework. 相似文献
In this paper, a dynamic modeling method and an active vibration control scheme for a smart flexible four-bar linkage mechanism featuring piezoelectric actuators and strain gauge sensors are presented. The dynamics of this smart mechanism is described by the Discrete Time Transfer Matrix Method of Multibody System (MS-DTTMM). Then a nonlinear fuzzy neural network control is employed to suppress the vibration of this smart mechanism. For improving the dynamic performance of the fuzzy neural network, a genetic algorithm based on the MS-DTTMM is designed offline to tune the initial parameters of the fuzzy neural network. The MS-DTTMM avoids the global dynamics equations of the system, which results in the matrices involved are always very small, so the computational efficiency of the dynamic analysis and control system optimization can be greatly improved. Formulations of the method as well as a numerical simulation are given to demonstrate the proposed dynamic method and control scheme. 相似文献
The objective of this study is to develop a tri-rotor flying robot, which adopts the Y-shaped three-rotor structure. In order to balance the yaw torque produced by the three rotors, the RC servomotor and linkage is installed on the tail axis in order to improve the angle of the rolling axis of the tail motor. Moreover, through the torque generated by the horizontal component of the lift from the inclined motor on the tail axis, it balances the yaw torque of the three rotors. The dynamic equations of the tri-rotor flying robot are determined in this paper. The relationship between motor thrust, angular acceleration, and voltage input was also studied in this research. In order to study the effect of control parameters on the flight stability completely, this study develops a universal stability experimental platform to help tuning the control parameters safely. Based on this, the tri-rotor flying robot can rapidly change flying gesture and avoid oscillation. Finally, we conducted some indoor and outdoor flight tests. From the experimental results, the tri-rotor flying robot can fly and hover stably in the sky. 相似文献
To overcome the disadvantages of the location algorithm based on received signal strength indication(RSSI) in the existing wireless sensor networks(WSNs),a novel adaptive cooperative location algorithm is proposed.To tolerate some minor errors in the information of node position,a reference anchor node is employed.On the other hand,Dixon method is used to remove the outliers of RSSI,the standard deviation threshold of RSSI and the learning model are put forward to reduce the ranging error of RSSI and improve the positioning precision effectively.Simulations are run to evaluate the performance of the algorithm.The results show that the proposed algorithm offers more precise location and better stability and robustness. 相似文献
In the presence of mixed micelle of a cationic and a nonionic surfactants,the reaction of scandium(Ⅲ)with phenylfluorone or its derivatives results in a very sensitive colouration:furthermore this chromophoricsystem tolerates the existence of a considerable amount of masking agents,so that many interfering ions can bemasked and its selectivity would be further improved.After studying the optimum conditions and main charac-ters of some similar systems,the system of Sc(Ⅲ)—Orthonitrophenyuorone—CetyltrimethylammoniumBromide(CTMAB)—Triton X-200 is chosen for the spectrophotometric determination of microamount ofscandium.The proposed method shows not only high sensitivity(with a molar absorptivity of 2.12×1~sLmol~(-1)·cm~(-1))and selectivity,but also high tolerance of chromophoric conditions.Beer's law is obeyed over arange of 0~6.0 μg Sc/25ml.If the dual-wavelength method is applied,a much higher molar absorptivity o|"3.04×10~5L·mol~(-1)·cm~(-1)than those of other systems ever reported in the literatures can be attained:more-over.its linear range is extended to 0~8.0 μg Sc/25ml and its reproducibility is also improved. 相似文献