雷单像素飞行目标分类通过质地融合轻量差异化操作员
Hubin Ling1,2, Bingzhang Hu2, Dongfeng Shi1,2
1Science Island Branch of Graduate School, University of Science and Technology of China, Hefei, 230026, China.
Scientific reports
|December 15, 2025
概括
这项研究引入了一种高效的深度学习模型,用于使用Radon单像素成像 (SPI) 在超低采样速率下对空中物体进行分类. 这种新的方法提高了无人机检测和安全应用的准确性.
科学领域:
- 计算机视觉 计算机视觉
- 信号处理 信号处理
- 机器学习 机器学习
背景情况:
- 单像素成像 (SPI) 为无人机监控和机场安全等应用提供快速采样和远程成像.
- SPI中的低采样率提高了速度,但降低了图像质量,挑战了对象识别.
- 现有的SPI分类方法使用浅层网络,限制它们对复杂任务的辨别能力.
研究的目的:
- 开发一个高效的深度学习模型,以超低的采样率对SPI进行分类.
- 为了提高在具有挑战性的成像条件下识别快速移动的空中物体的准确性.
- 为了利用Radon SPI特性,并将传统的图像处理技术集成到深度学习框架中.
主要方法:
- 使用最先进的轻量级分类模型作为基础.
- 集成传统的纹理和线程过运算符到可区分的模块中.
- 专门为SPI数据设计和优化了一种新型分类模型.
主要成果:
- 拟议的模型在定制的Radon SPI飞行目标数据集上实现了最高的Top-1精度.
- 与现有的最先进的轻量级分类模型相比,表现出卓越的性能.
- 验证了将Radon SPI特征的先前知识整合到深度学习中的有效性.
结论:
- 先进的深度学习模型,以特定领域的特征进行优化,可以克服Radon SPI.低采样率的局限性.
- 开发的模型显示了在安全和监控应用中增强空中物体分类的重大前景.
- 源代码将公开发布,以促进进一步的研究和开发.
更多相关视频
09:21Human Brown Adipose Tissue Depots Automatically Segmented by Positron Emission Tomography/Computed Tomography and Registered Magnetic Resonance Images
Published on: February 18, 2015
12.6K
03:31Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
Published on: December 15, 2023
991
相关概念视频
01:15Determination of Pi Terms
The Buckingham Pi theorem is a valuable method in dimensional analysis, reducing complex relationships between variables into dimensionless terms. Relevant variables in analyzing the lift force on an airplane wing include lift force, air density, wing area, aircraft velocity, and air viscosity. Expressing each variable in terms of fundamental dimensions — mass, length, and time — provides a consistent foundation for constructing these dimensionless terms.
The theorem indicates that the number...
The theorem indicates that the number...
01:29Real-World Applications of Space Curves
Modern aerospace navigation depends on the accurate prediction of motion in three-dimensional space. In defense applications, radar systems continuously track both interceptors and moving aerial targets to find whether their flight paths will result in a collision. These motions are modeled mathematically as space curves, which represent paths that change continuously with time. Each object’s position is described by a vector function that specifies its location in terms of time-dependent...
01:30Vector Functions and Motion: Problem Solving
Accurate position tracking is fundamental to the safe and effective operation of unmanned aerial vehicles (UAVs), particularly during precision maneuvers near complex structures. In this scenario, a drone is programmed to perform a high-precision inspection of a vertical structure, starting at position ((x, y, z) = (3, 0, 0)), with an initial velocity oriented in the positive z-direction. The trajectory of the drone is governed by a time-dependent acceleration function a(t), which is predefined...