No abstract

Based on the two-dimensional (2D) discrete Rulkov model that is used to describe neuron dynamics, this paper presents a continuous non-autonomous memristive Rulkov model. The effects of electromagnetic induction and external stimulus are simultaneously considered herein. The electromagnetic induction flow is imitated by the generated current from a flux-controlled memristor and the external stimulus is injected using a sinusoidal current. Thus, the presented model possesses a line equilibrium set evolving over the time. The equilibrium set and their stability distributions are numerically simulated and qualitatively analyzed. Afterwards, numerical simulations are executed to explore the dynamical behaviors associated to the electromagnetic induction, external stimulus, and initial conditions. Interestingly, the initial conditions dependent extreme multistability is elaborately disclosed in the continuous non-autonomous memristive Rulkov model. Furthermore, an analog circuit of the proposed model is implemented, upon which the hardware experiment is executed to verify the numerically simulated extreme multistability. The extreme multistability is numerically revealed and experimentally confirmed in this paper, which can widen the future engineering employment of the Rulkov model.

No abstract

Object detection for remote sensing images has problems such as complex backgrounds, multi-scale and difficulty in the detection of small targets. Because of the above problems, an improved object detection algorithm for remote-sensing images is proposed. Firstly, fusing the characteristics of shallow and deep networks, the original feature pyramid structure is reconstructed, the improved adaptive feature fusion structure is introduced before the network prediction, and the location and category information of different feature maps are fused. Secondly, the attention mechanism is introduced into the network to reduce the interference caused by complex backgrounds; Finally, the up-sampling module is improved to expand the receptive field and realize up-sampling based on semantic information. Through experiments, we show that our network achieves 93.9% and 98.5% detection accuracy on Levir and RSOD dataset, respectively, with the detection speed, reaching 98.26fps. Compared with the original network, the accuracy is increasing by 4% and 1.2%, respectively. Experimental results show that the detection accuracy of our algorithm is improved while maintaining the light weights and efficiencies of the original network.

No abstract