一个高效的重定型小物体检测变压器用于热红外图像.
Canhao Guo1,2, Peidong Luo1,2, Zhixing Ma1,2
1School of Artificial Intelligence, Shenzhen Technology University, Shenzhen, 518118, China.
Scientific reports
|December 24, 2025
概括
本研究介绍了PWL-RTDETR,这是一个高效的基于变压器的框架,用于在红外图像中检测小物体. 它实现了高精度并降低了计算负载,使其成为无人机 (UAV) 等资源有限平台的实时部署的理想选择.
科学领域:
- 计算机视觉 计算机视觉
- 人工智能的人工智能
- 机器学习 机器学习
背景情况:
- 在热红外图像中检测小物体是具有挑战性的,因为对比度低,纹理有限.
- 在无人机 (UAV) 等边缘平台上部署面临着计算限制.
研究的目的:
- 为红外小型物体检测开发一个高效的基于变压器的框架.
- 解决在资源有限的平台上实时部署的计算限制.
主要方法:
- 拟议的PWL-RTDETR框架具有部分卷积修复参数化块 (PConvRep-Block) 进行高效的计算.
- 推出了WTRCSPNeck,这是一个轻量级的架构,与CNCSPELAN和WTConv一起用于增强的多尺度特征聚合.
- 实现了基于尺度的层适应性修剪,用于模型压缩和散散.
主要成果:
- 与HIT-UAV和LLVIP红外数据集上最先进的模型相比,PWL-RTDETR显示出更高的准确性.
- 在模型参数和浮点运算 (FLOP) 中实现了显著的减少.
- 该模型被证明适用于资源有限的环境中的实时应用.
结论:
- PWL-RTDETR提供了一个有效的解决方案,用于在红外图像中准确有效地检测小物体.
- 该框架的效率和准确性使其适用于无人机等边缘设备上的实时感知任务.
相关概念视频
Transformers with Off-Nominal Turns Ratios
490
In scenarios involving parallel transformers with disparate ratings, developing per-unit models requires accommodating off-nominal turns ratios. This situation arises when the selected base voltages are not proportional to the transformer’s voltage ratings. Consider a transformer where the rated voltages are related by the term a. If the chosen voltage bases satisfy a relationship involving term b, term c is defined as the ratio of these bases. This ratio is then substituted into the...
490
The Ideal Transformer
1.3K
In single-phase two-winding transformers, two windings are coiled around a magnetic core characterized by cross-sectional area A and magnetic permeability μ. A phasor current i1 enters the left winding while i2 exits the right winding, establishing the fundamental working of the transformer through electromagnetic principles.
Ampere's Law forms the basis of understanding the magnetic field within the transformer. It states that the integral of the magnetic field intensity's tangential...
Ampere's Law forms the basis of understanding the magnetic field within the transformer. It states that the integral of the magnetic field intensity's tangential...
1.3K
