Object tracking in satellite videos faces various challenges such as target occlusion, rotation, and background clutter. This study proposes a correlation particle filter algorithm with motion estimation for object videos. The tracker, called CPKF, combines the strengths of correlation, particle, Kalman filters. Compared existing methods based on filters, proposed tracker has three major advantages: (1) Particle sampling, build robustness against partial complete occlusion. (2) Color...

Recent deep learning based-works have made remarkable progress in Remote-Sensing Image Super-Resolution (RSISR). However, the complicated network architecture as well a huge amount of parameters increase computational cost, hindering their practical deployment. To alleviate this problem, we propose novel Re-parameterized Feature Distillation Network (ReFDN) for lightweight and efficient RSISR tasks. distillation, refinement, condensation, enhancement are efficiently integrated into...

Fine-grained ship classification (FGSCR) has many applications in military and civilian fields. In recent years, deep learning been widely used for tasks, its success is inseparable from that of big data. However, images are valuable, with only a few specific category being obtained, leading to the fine-grained few-shot problem. addition, feature map channels contain distinct characteristics discriminative details, which significantly influence FGSCR. Intuitively, should be assigned larger...

Red mud was used in the supercritical water oxidation (SCWO) disposal of sewage sludge, not only as a neutralizer for acidic substances produced <italic>in situ</italic>, but also catalyst decomposition pollutants.

In this letter, a stationary-wavelet-based despeckling algorithm based on the two-sided generalized gamma distribution (GΓD) model is proposed. We first introduce GΓD as flexible and efficient for wavelet coefficients of logarithmically transformed synthetic aperture radar intensity or amplitude. The strength highlighted in terms its fit to data, low computational cost, ease parameter estimation. By empirical results, we then motivate noise-free signal. parameters are estimated with moment...

Ship detection with several military and civilian applications has drawn considerable attention in recent years. In remote sensing images, ships have the characteristics of arbitrary orientation. Based on characteristics, many arbitrary-oriented ship detectors been proposed. Most these preset horizontal or rotated anchors determine positive negative samples based intersection over union (IoU) between anchor ground-truth bounding box, what is called label assignment process. However, IoU...

In synthetic aperture radar (SAR) images, the edge of shadow is blurred because moving while data are collected. this letter, problem expanded on by using imaging formation perspective. First, an approximate method to represent quality boundary region based quadratic phase errors (QPEs) provided for first time, which built up relationship between parameters caster and behavior in SAR image. We notice that QPE approximately a linear function height caster. Second, we deduced height-dependent...

Cylinder tanks appear as arcs in SAR image. It is found the radius of some bigger than tank. To analysis this phenomenon, combining scattering theory and principle SAR, a novel prediction model about image cylinder tanks, including with fixed top floating top, proposed. Analysis on details provided by model. Several experiments are implemented.

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...

In this paper, we propose a novel radar imaging simulator with precise simulation of geometry and relative radiometry, to assist target recognition. The is based on an new visualization procedure Lambertian-specular mixture model. Simulated results are compared MiniSAR image airplane C-130 evaluating the efficiency.

The super-resolution (SR) reconstruction is drawing increasing attention in remote-sensing image processing, owing to improving the spatial resolution and enriching details of initially obtaining low-resolution (LR) images. Different from natural images, remote sensing images usually have more complex distribution degradation processes. Although current deep convolutional neural network (DCNN)-based approaches reach better performance by deepening introducing mechanism, this accompanied...

The target of this paper is to cultivate students' innovative ability.According data structure teaching process, proposed design concept double mainline firstly.Then, in order reflect the thinking course, constructed three-layer framework which included content, mode and means.This also 4 level staged practice system contain experiment teaching, course design, innovation contest scientific research project development.

This paper proposes a novel radar imaging simulator based on analytical electromagnetic and geometric models for common typical kind of urban structure. The are presented not only improving the efficiency simulation but also direct further understanding features in SAR image.