AbstractDeep learning has shown great power in processing remote sensing data, especially for fine-grained ship image classification. However, the lack of a large amount effective training data greatly limits performance neural networks. Based on current augmentation methods, images ships sea generated have problem distortion, blurring, and poor diversity. To tackle this problem, we propose novel progressive method that combines simulation samples style transfer (NST) based network to...

The flexible job shop scheduling problem has always been the focus of research in manufacturing field. However, most previous studies focused more on efficiency and ignored energy consumption. Energy, especially non-renewable energy, is an essential factor affecting sustainable development a country. To this end, paper designs model with consideration line production Except for processing stage, consumption transport, set up, unload, idle stage are also included our model. weight property...

In this article, analysis has been made with regard to current situation of ecological revetment works and requirements river channels in the middle lower reaches Yangtze River, types etc. The development trend its application prospect are also discussed. indicates that, after impoundment Three Gorge Reservoir, erosion River will result severe like local channel adjustment bank collapse continuous weakening system. It is necessary keep on strengthening comprehensive control adopting future.

Ship target detection faces the challenges of complex and changing environments combined with varied characteristics ship targets. In practical applications, complexity meteorological conditions, uncertainty lighting, diversity can affect accuracy efficiency algorithms. Most existing methods perform well in conditions a general scenario but underperform conditions. this study, collaborative network for under foggy weather is proposed, aiming to achieve improved while satisfying need...

A slowdown subduction model is presented to estimate slab breakoff depth and initial time when continental lithosphere subducts into mantle. By assuming oceanic with uniform velocity (all forces acting on plate are balanced) noticing that, leading order, some resistant proportional functions of velocity, a simple dynamic proposed describe the retarded process. Thermal strength structures within calculated under this We then compare integrated transitional zone from buoyancy change due...

Aircraft classifiers in remotely sensed images based on deep convolutional neural networks play a significant role military. However, practical applications, there is lack of remote sensing fine-grained aircraft data. In this study, we demonstrate that few-shot learning (FSL) can be effectively used for identification and propose new classifier-adaptive earth mover's distance (Adap-EMD) recognition few-sample aircraft. Adap-EMD consists an efficient block attention mechanism (EBAM) adaptive...

Remote sensing image ship fine-grained classification is a challenging vision problem due to factors such as inter-class similarity, real-world scarcity, and class imbalance. Data augmentation aims solve these problems from the data perspective. Most of methods cannot work effectively in complex scenes, which restricts their practical application. Here, we propose novel pipeline based on local-aware translation achieve representation mapping between cross-domain corresponding instances. Our...

The availability of many commercial satellites has created favorable conditions for tracking typical objects in remote sensing sequences, making them widely useful numerous applications. However, small objects, multiple similar disruptors, background clutter, and occlusion are significant challenges to this field. This study proposes the novel tracker-temporal motion compensation Siamese network (Siam-TMC) tracking. Our method relies on a multidimensional information-aware module temporal...

摘要: 面向天文观测的空间科学卫星任务规划是一个复杂的多目标优化问题.通过对天文观测类卫星的任务规划要素及约束条件进行抽象,建立了面向天文观测的多目标任务规划问题模型,在此基础上设计了基于NSGA-II的多目标观测任务规划算法,并通过实例进行了实验及结果分析.研究表明,该方法能够有效解决天文观测类卫星不同规模的任务规划问题. 关键词: 空间科学卫星 / 任务规划 多目标进化算法

The effect of the atmosphere on propagating energy during remote sensing imaging simulation is one most critical factors affecting quality image. classical methods retrieving atmospheric optical parameters (AOP) have shortcomings in GPU-based application. This paper proposed a method based deep neural networks (DNN) and principal component analysis (PCA) to compute AOP. Firstly, MODTRAN employed obtain large numbers AOP as original spectrum set different weather observation geometry...

Interpreting airborne remote sensing images plays an important role in aviation control and battlefield situational awareness. However, highly dynamic aircraft detection remains challenging, owing to variable object sizes, flexible attitudes, motion blur. This study develops a multi-scale dataset benchmark overcome challenges, such as high intraclass variance, multiple scales angles, blur, partial occlusion. We also propose trained-from-scratch detector, the local feature correlation single...

Linear and nonlinear counter-traveling waves in a fluid-filled annular cylinder with realistic no-slip boundary conditions uniformly heated from below rotating about vertical axis are investigated. When the gap of is moderate, there exist two three-dimensional traveling driven by convective instabilities: retrograde mode localized near outer sidewall prograde adjacent to inner different wave number, frequency critical Rayleigh number. It found that retrogradely propagating always more...

To address the problem of intelligent recognition optical ship targets under low-altitude squint detection, we propose an method based on simulation samples. This comprehensively considers geometric and spectral characteristics ocean background performs full link modeling combined with detection atmospheric transmission model. It also generates expands multi-angle imaging samples in visible light band using expanded sample type to perform feature analysis modification SqueezeNet. Shallow...

In recent years, due to deteriorating environment, coal-to-electricity project has received a lot of attention. However, is still in the early stage development, and relevant standards electric heating load connecting distribution network are not perfect. Based on current problems project, order improve renewable energy generation (REG) integration capacities network, typical construction schemes put forward this paper. Considering reliability economic efficiency, two-step scheme evaluation...

Introduction: Log anomaly detection is essential for monitoring and maintaining the normal operation of systems. With rapid development maturation deep learning technologies, learning-based log has become a prominent research area. However, existing methods primarily concentrate on directly detecting data in single stage using specific information, such as sequential information or semantic information. This leads to limited understanding data, resulting low accuracy poor model robustness....

Abstract With the continuous advancement of technology, intelligent solutions have found widespread applications in various fields, including pump stations water management projects. This paper focuses on key technologies involved station monitoring, control, and operation. The takes a Station as an example addresses problems current situation its management. It designs architecture at three layers: business application, technical methods, requirements. Solutions are proposed for dispatch...

Linear and nonlinear convection in a rotating annular cylinder, under experimental boundary conditions, heated from below about vertical axis are investigated. In addition to the usual physical parameters such as Rayleigh Taylor number, an important geometric parameter, ratio of inner outer radius, enters into problem. For intermediate ratios, linear stability analysis reveals that there exist two countertraveling convective waves which nonlinearly significant: retrograde wave located near...

Hyperspectral images (HSIs) are widely used as data formats for remote sensing. Correspondingly, the spectral super-resolution (SSR) technique to generate high-spatial-resolution HSIs from sensing multispectral (MSIs) has emerged a popular research topic owing its high costs and hardware requirements. Generally, existing SSR methods obtain an HSI MSI of natural scenes. However, these barely learn complex spectra found in lack effective treatments degradation phenomenon In this study,...

Remote sensing images (RSIs) are limited by low-resolution imaging and a variety of degradation effects, posing significant challenges for high-level visual tasks. Traditional deep learning-based super-resolution (SR) algorithms struggle to optimize these issues end-to-end, which inhibits the algorithm's lightweight design restricts its application in real-time scenarios. To address this, we introduce multi-branch attention integration network (MAIN) degraded RSI SR. This features two key...