基于深度学习的自动驾驶汽车对道路裂的视觉检测
Ibrahim Meftah1, Junping Hu1, Mohammed A Asham2
1College of Mechanical and Electrical Engineering, Central South University, Changsha 410017, China.
Sensors (Basel, Switzerland)
|March 13, 2024
概括
本研究提出了一种有效的方法来检测使用深 convolutional 神经网络 (CNN) 与随机森林结合使用的道路裂. 该方法在从图像中识别路面断裂方面取得了高准确性.
科学领域:
- 土木工程 土木工程是指土木工程.
- 计算机科学 计算机科学
- 人工智能的人工智能
背景情况:
- 道路裂检测对于评估混凝土路面完整性至关重要.
- 传统方法与杂的表面和现实世界的条件作斗争,影响自动驾驶汽车的安全性.
- 开发强大的,自动化裂检测对于基础设施维护至关重要.
研究的目的:
- 引入一种基于图像的先进道路裂检测方法.
- 将随机森林与深层卷积神经网络 (CNN) 结合起来,以提高准确性.
- 为了评估最先进的CNN模型在识别混凝土路面裂方面的性能.
主要方法:
- 使用了三种深度CNN模型:移动网络,InceptionV3和Xception.
- 在3万张图像的数据集上训练模型,开发出有效的裂检测系统.
- 通过系统地比较不同基础学习率的验证准确性来优化模型性能,确定0.001是最佳的.
主要成果:
- 实现了最大的验证准确率为99.97%,最佳基础学习率为0.001.
- 在6000张未见的测试图像 (224x224像素) 上评估受过训练的模型.
- 证明了出色的测试性能,准确率为99.95%,精度为99.95%,回忆率为99.94%,F1分数为99.94%.
结论:
- 拟议的混合方法有效地检测在真实混凝土表面上的道路裂.
- 深度CNN模型,特别是当优化时,为路面检查提供了强大而灵活的解决方案.
- 这种技术在通过自动化裂识别来提高道路安全和维护方面具有重大前景.
相关概念视频
01:20Microcracking in Concrete
117
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...
117
01:29Sight Distance in a Vertical Curve
45
Sight distance on vertical curves is critical in roadway design. It ensures drivers can see far enough ahead to identify and respond to hazards effectively. This directly impacts safety, driver comfort, and the overall efficiency of the transportation network.Vertical curves are classified into crest and sag curves based on their geometry. For crest curves, sight distance is determined by the line of sight between a driver's eye and a small object on the road's surface. Design parameters for...
45