实时无人机图像对象检测的增强功能表示,使用上下文信息和自适应融合
Junbao Wu1,2, Hao Meng3,4, Ming Yuan1
1College of Intelligent Systems Science and Engineering, Harbin Engineering University, Nantong Street, Harbin, 150001, China.
Scientific reports
|September 29, 2025
概括
本研究介绍了YOLO-UD,这是一种用于无人机 (UAV) 图像的增强实时物体检测网络. 通过整合上下文信息和自适应的多尺度融合,YOLO-UD提高了准确性和速度,以更好地检测小物体.
科学领域:
- 计算机视觉 计算机视觉
- 人工智能的人工智能
- 机器人技术 机器人技术 机器人技术
背景情况:
- 在无人机 (UAV) 图像中实时对象检测面临诸如小物体,遮蔽和不均分布等挑战.
- 现有的算法,包括YOLO变体,当直接应用于无人机数据集时,显示性能下降.
- 目前的解决方案缺乏对现实世界无人机部署场景的全面方法.
研究的目的:
- 开发一个增强的实时物体检测网络 (YOLO-UD),以提高无人机图像的性能.
- 为了应对在无人机数据中检测小物体,遮蔽和不同目标分布的具体挑战.
- 为了实现基于无人机的下游任务的精度和推断速度之间的卓越平衡.
主要方法:
- 推出了YOLO-UD,基于YOLO11架构构建,结合了一种新的C3kHR模块,具有扩展卷积,用于多尺度特征表示.
- 设计了一个高效的自适应特征融合网络 (EAFN),以过和优先考虑多尺度特征信息.
- 集成了一个小物体检测层 (SMDL),专门增强小目标的检测.
主要成果:
- 在无人机图像上,YOLO-UD在实时物体检测准确性和速度方面取得了显著的改进.
- C3kHR模块有效地捕获了丰富的上下文和多尺度特征.
- 欧洲农村网络成功过并对相关信息进行优先排序,而SMDL提高了小物体检测能力.
结论:
- 在具有挑战性的无人机环境中,YOLO-UD为实时物体检测提供了强大而有效的解决方案.
- 拟议的网络在准确性和推断速度之间实现了强大的平衡,优于现有方法.
- 这些发现验证了整合上下文信息和适应融合用于无人机图像分析的有效性.
相关概念视频
01:21Association Areas of the Cortex
8.9K
Association areas are regions of the cerebral cortex that do not have a specific sensory or motor function. Instead, they integrate and interpret information from various sources to enable higher cognitive processes such as memory, learning, and decision-making. Some key association areas include the following:
Prefrontal Association Area: This area is located in the frontal lobe and is involved in planning, decision-making, and moderating social behavior. It connects with primary motor areas,...
Prefrontal Association Area: This area is located in the frontal lobe and is involved in planning, decision-making, and moderating social behavior. It connects with primary motor areas,...
8.9K
01:22Force Classification
2.3K
Forces play a crucial role in the study of physics and engineering. They are essential in describing the motion, behavior, and equilibrium of objects in the physical world. Forces can be classified based on their origin, type, and direction of action.
Contact and non-contact forces are two of the most widely used categories of forces. As the name suggests, contact forces require physical contact between two objects to act upon each other. Examples of contact forces include frictional,...
Contact and non-contact forces are two of the most widely used categories of forces. As the name suggests, contact forces require physical contact between two objects to act upon each other. Examples of contact forces include frictional,...
2.3K