The classification of highly complex and detail-rich images, which is essential for botanical research and environmental monitoring, presents a significant challenge in electron microscope pollen analysis due to the need to enhance both accuracy and computational efficiency. This study introduces the SNT model, In order to achieve this, it incorporates Large-view Spatial Bilateral Attention (LSA), which is adept at capturing both local details and global contextual information. A novel lightweight pruning technique is employed to retain essential weight information while significantly reducing the model’s size, thereby enhancing its utility and efficiency. The results of our experimental study demonstrate that the SNT model exhibits a significant enhancement in accuracy compared to the advanced CNNs and Transformer model.

Problems of large sizes and low displacement resolutions widely exist in most of the pole-climbing robots based on electromagnetic or electrostatic motors, which greatly decrease the reachable workspace and influence their utility for inspection tasks. Here, we present a piezoelectric pole-climbing robot (Crabbot) to overcome these shortcomings, which has the advantages of small size, high load-to-weight ratio and high resolution. Moreover, Crabbot can climb poles with different cross-sections through the designed crab-like clamp. This robot was composed of two clamping units and an elongation unit, and the stepping displacements were amplified by the designed flextensional mechanism. A prototype with weight of 0.35 kg was manufactured, the external dimension was 50 × 54 × 86 mm. Experimental results showed that the robot achieved a maximum speed of 100.9μm/s under a voltage of 150 V and a frequency of 9 Hz; moreover, a load weight ratio of 1.42 and a displacement resolution of 85 nm were obtained. As a demonstration of its potential for industrial applications, Crabbot achieved stable step motion on climbing poles of different cross-sections under different climbing angles.

ABSTRACT This paper addresses the issues of rate control and routing for cloud data center networks. Based on the theory of the supply–demand equilibrium, we propose a fixed point model for formulating cloud network equilibrium problems in which the equilibrium conditions are given by nonlinear equations. We show that the network equilibrium point is the optimal solution of a nonlinear programming problem by utilizing the tools of the variational inequality and convex optimization. The augmented Lagrangian multiplier algorithm is used to solve the nonlinear programming problem for computing the network equilibrium point. Further consideration is given to the equilibrium problems of a cloud network with multirate multicast sessions. We evaluate our approach on some random networks with unicast and multicast sessions, and the results show the effectiveness of our approach in finding the optimal equilibrium rates. We further evaluate the performance of our approach through cloud data center network simulation under various parameter settings, and the results show that the performance of our algorithm can be tuned and improved by choosing appropriate parameter values. Copyright © 2013 John Wiley & Sons, Ltd.

We consider the problem of elastic-demand network equilibrium in communication networks which support both multirate multicast sessions and unicast sessions. Extending existing work for unicast sessions, we first present the elastic-demand network equilibrium models in different multicast sessions, which can be formulated as a convex programming problem. To solve the convex programming problem we use the augmented Lagrangian multiplier algorithm in which the attractive features of the exterior penalty with primal-dual methods and Lagrangian multipliers concepts are combined while curtailing the disadvantage of both. Some numerical results about unicast sessions and multirate multicast sessions are demonstrated through efficient implementations of the augmented Lagrangian multiplier algorithm.