Visual object tracking has been a concern topic these years, and many trackers have achieved good results in various fields. These researches and breakthroughs have made many improvements to solve problems such as dri...
Visual object tracking has been a concern topic these years, and many trackers have achieved good results in various fields. These researches and breakthroughs have made many improvements to solve problems such as drift, lighting, deformation and occlusion. In this paper, we improve the structure of the AlexNet[1] network by designing the three important influencing factors of the receptive field size, total network step size, and feature filling of the twin network. Apart from this, we add a smoothing matrices and a background suppression matrices to effectively learn the features of the first few frames as much as possible. Fuse multilayer feature elements can learn online about target appearance changes and background suppression, and we train them by using continuous video sequences.
As the key basis materials of Ni-MH and Ni/Cd battery electrode, the quality of the nickel foam products is closely related with the performance and safety of rechargeable batteries. In order to meet the requirements ...
详细信息
The rapid development of autonomous driving in recent years presents lots of challenges for scene understanding. As an essential step towards scene understanding, semantic segmentation thus received lots of attention ...
详细信息
Gastric cancer is the second leading cause of cancer‐related deaths worldwide, and the major hurdle in biomedical image analysis is the determination of the cancer extent. This assignment has high clinical relevance ...
详细信息
This paper presents an on-line estimator that incorporates adaptive MIMO radical basis function neural networks (RBFNNs) for model identification of quadrotor unmanned aerial vehicles (UAVs). The inputs and outputs of...
详细信息
ISBN:
(纸本)9781509061914
This paper presents an on-line estimator that incorporates adaptive MIMO radical basis function neural networks (RBFNNs) for model identification of quadrotor unmanned aerial vehicles (UAVs). The inputs and outputs of quadrotor aircrafts can be obtained from dynamic models or real attitude and position sensors. The adaptive learning rate is employed in the gradient descent method for the update of the weights of RBFNNs, and Lyapunov approach guarantees the stability of the global convergence of the modeling errors. The Welsch functions are also employed as the error functions to get rid of the influence from the noise due to disturbances like wind gusts. Simulation results using robotics Toolbox for Matlab verify the effectiveness and robustness of the proposed estimator compared with results of traditional RBFNNs. Experiment results from real aircraft platform show that RBFNNs combining adaptive learning rate and Welsch error functions can approximate the overall system with high accuracy and robustness to disturbances.
暂无评论