相关实验视频

Updated: Jun 29, 2025

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

一种深度学习方法用于在无人机辅助基础设施检查中对表面裂纹进行分类和细分

Shamendra Egodawela¹, Amirali Khodadadian Gostar¹, H A D Samith Buddika²

¹School of Engineering, RMIT University, 124 La Trobe St, Melbourne, VIC 3000, Australia.

Sensors (Basel, Switzerland)

|March 28, 2024

概括

这项研究引入了一个快速地表裂检测系统,使用两个无人机 (UAV) 和一个名为CrackClassCNN的新型卷积神经网络 (CNN). 该系统实现了95.02%的准确性,大大提高了基础设施检查效率.

科学领域:

土木工程土木工程是指土木工程.
计算机视觉计算机视觉
机器人技术机器人技术机器人技术

背景情况:

表面裂检测对于基础设施健康监测至关重要.
传统的检查方法耗时且劳动密集.
进入基础设施中的狭窄空间带来了重大挑战.

研究的目的:

开发一个快速可靠的系统,用于基础设施的表面裂检测.
评估一种新的卷积神经网络 (CNN) 架构的有效性,CrackClassCNN,用于裂分类.
评估分段任何模型 (SAM) 对于裂纹细分的性能.

主要方法:

部署两个无人驾驶飞行器 (UAV) 以同时获取图像.
使用二进制分类CNN (CrackClassCNN) 来识别图像中的裂.
采用分段任何模型 (SAM) 来对检测到的裂区域进行细分.
基准测试CrackClassCNN与最先进的CNN架构相比,SAM与手动注释相比.

主要成果:

小说CrackClassCNN的分类准确率达到了95.02%.
细分任何模型 (SAM) 在裂细分方面表现出强的表现,平均IOU为0.778和F1得分为0.735.
集成的无人机和CNN系统被证明对高效的基础设施检查非常有效.

结论:

开发的无人机平台和CrackClassCNN为快速和可靠的表面裂检测提供了变革性的解决方案.
该系统特别适合在狭窄的空间中检查基础设施.
先进的无人机技术和人工智能驱动的图像分析的结合大大提高了基础设施健康调查.

关键词:

混凝土裂是混凝土上的裂.卷积神经网络 (CNN) 是一种神经网络.深度学习是一种深度学习.无人驾驶飞行器 (UAV) 是一种无人驾驶飞行器.

更多相关视频

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

相关概念视频

Microcracking in Concrete

Microcracking in Concrete

Microcracking in concrete refers to the tiny cracks that can form within the material even before any external load is applied. These microcracks typically occur at the interface between the coarse aggregate and the hydrated cement paste, often as a result of differential volume changes prompted by variations in stress-strain behavior, as well as thermal and moisture movement. Initially, these microcracks remain stable and do not grow substantially until the concrete is stressed to about 30...