The term fractal is used to describe an object displaying self-similarity at different scales. This can be measured by either the power-law exponent or ht-index, which a recently proposed method for characterizing nature of geographic features. Although increasingly popular in geography, ht-index not sensitive changes evolutions fractals, limiting its usefulness as alternative "fractal dimension" exponent. Two improvements were suggested literature, namely, cumulative rate growth (the CRG...

Due to the change of flight altitude and attitude UAV, object scale in UAV images exists difference which leads a great challenge for detection has drawn wide attention. In this paper, an improved network named ECascade-RCNN is proposed deal with multi-scale problem task images. We present innovative Trident-FPN backbone extract features design new attention mechanism enhance performance detector. Moreover, k-means algorithm adapted generate anchors so that model can get better regression...

Although the existing semantic image synthesis methods based on generative adversarial networks (GANs) have achieved great success, quality of generated images still cannot achieve satisfactory results. This is mainly caused by two reasons. One reason that information in layout sparse. Another a single constraint effectively control position relationship between objects image. To address above problems, we propose dual-conditional GAN with an external attention for (DCSIS). In DCSIS,...

The rapid development of the Internet has broadened channels dissemination information, it also led to and widespread propagation rumors, which can have a serious negative impact socially. In this paper, an improved ISR-WV rumor model integrating multichannels is proposed by considering system’s time delay, influence different on dynamic process further analyzed. Moreover, basic reproduction number R 0 , rumor-free equilibrium, rumor-prevailing as well their stability, are deduced. Then,...

The wide spread of rumor is undoubtedly harmful to social stability; we should try lower the effect on society. Therefore, it reasonable put forward control strategy basis study law propagation. Firstly, ISTR model established by including influencing factors true information spreader and reinforcement. And using next generation matrix method, basic reproduction number obtained. Then, in order minimize adverse effects rumors, through introducing two strategies scientific knowledge...

With the development of science and technology, products are being updated more quickly. Therefore, diffusion product information can make people better choose products. It is very meaningful to study competition multiple information. In this paper, dual model with preference was proposed based on mean-field equation in complex networks. The dynamic analyzed by analysis Markov chains. According Monte Carlo simulation mechanism, critical threshold obtained. accuracy theory results verified...

Monocular 3D detection is to obtain the information of object from image. The mainstream methods mainly use L1 loss or L1-like control instance depth prediction. However, these have not achieved satisfactory results. One main reasons that does accurately reflect fit between predicted and corresponding ground truth. Another reason on RGB image hard be directly learned in network. In order solve above problems, a novel thermodynamic based principle free energy minimisation decoupling method...

Abstract The rapid development of the Internet has accelerated spread rumors, posing challenges to social cohesion and stability. To address this, a multi-channel rumor propagation model incorporating individual game behavior time delay is proposed. It depicts individuals strategically choosing channels in process, capturing real-world intricacies more faithfully. Specifically, allowing spreaders choose between text video information base channels. Strategy adoption hinges on benefits versus...

In the scenario of Unmanned Aerial Vehicle (UAV), angle and height at which UAV observes object cause partial occlusion, deformation, size changing image. Based on Discriminative Correlation Filter (DCF) algorithm, this paper proposes a new tracking algorithm DCF-GA (Discriminative with Generation Adversarial example) to achieve anti-occlusion in UAV. Firstly, we design mask selection strategy generate adversarial example can enhance antiocclusion performance our algorithm. The response...

Landmark detection and recognition algorithm is a very important technology for vision-based Unmanned Aerial Vehicles (UAVs) autonomous pitching. The deformation rotation of landmarks the background distraction will be challenges recognition. Based on Support Vector Machine (SVM) appearance features landmarks, landmark proposed in this paper. presents scheme based ellipse which forms ellipses by optimized arcs estimates parameters decomposed space using Hough transform. To get better edge...