Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Pavement crack detection and small surface element fitting detection method based on YOLOv5
The invention discloses a pavement crack detection and small surface element fitting detection method based on YOLOv5. The method comprises the following steps: acquiring a target detection data set; preprocessing the image; classifying and calibrating the image data; dividing a data set; establishing a network and carrying out detection; a G-YOLOv5 network model is trained; setting a loss function; a G-YOLOv5 network model test is carried out; calibrating a facet metadata set; dividing a facet metadata set; training a VIT network model; testing a VIT network model; and calculating the crack area. According to the pavement crack detection and small surface element fitting detection method based on YOLOv5, the problem that a traditional target detection algorithm cannot accurately recognize the defect area of a pavement crack can be solved, calculation of the invalid area of the crack defect is reduced, and the recognition capacity of small cracks and fuzzy cracks is improved; the crack area is divided carefully, the precision of the crack area can be improved to a greater extent through block fitting, calculation errors are reduced, and therefore the finer crack area is obtained.
本发明公开了一种基于YOLOv5的路面裂缝检测及小面元拟合检测方法，包括获取目标检测数据集；图像预处理；图像数据分类标定；数据集划分；搭建网络并进行检测；G‑YOLOv5网络模型训练；损失函数设置；G‑YOLOv5网络模型测试；标定小面元数据集；小面元数据集划分；VIT网络模型训练；VIT网络模型测试；裂缝面积计算。本发明的基于YOLOv5的路面裂缝检测及小面元拟合检测方法能够解决传统的目标检测算法不能准确识别路面裂缝缺陷面积的问题，减少了对裂缝缺陷无效面积的计算，提高了对细小裂缝、模糊裂缝的识别能力；对裂缝区域进行细致的划分，分块拟合可以更大程度的提高裂缝面积的精度，减少计算误差，从而得到更精细的裂缝区域面积。
Pavement crack detection and small surface element fitting detection method based on YOLOv5
The invention discloses a pavement crack detection and small surface element fitting detection method based on YOLOv5. The method comprises the following steps: acquiring a target detection data set; preprocessing the image; classifying and calibrating the image data; dividing a data set; establishing a network and carrying out detection; a G-YOLOv5 network model is trained; setting a loss function; a G-YOLOv5 network model test is carried out; calibrating a facet metadata set; dividing a facet metadata set; training a VIT network model; testing a VIT network model; and calculating the crack area. According to the pavement crack detection and small surface element fitting detection method based on YOLOv5, the problem that a traditional target detection algorithm cannot accurately recognize the defect area of a pavement crack can be solved, calculation of the invalid area of the crack defect is reduced, and the recognition capacity of small cracks and fuzzy cracks is improved; the crack area is divided carefully, the precision of the crack area can be improved to a greater extent through block fitting, calculation errors are reduced, and therefore the finer crack area is obtained.
本发明公开了一种基于YOLOv5的路面裂缝检测及小面元拟合检测方法，包括获取目标检测数据集；图像预处理；图像数据分类标定；数据集划分；搭建网络并进行检测；G‑YOLOv5网络模型训练；损失函数设置；G‑YOLOv5网络模型测试；标定小面元数据集；小面元数据集划分；VIT网络模型训练；VIT网络模型测试；裂缝面积计算。本发明的基于YOLOv5的路面裂缝检测及小面元拟合检测方法能够解决传统的目标检测算法不能准确识别路面裂缝缺陷面积的问题，减少了对裂缝缺陷无效面积的计算，提高了对细小裂缝、模糊裂缝的识别能力；对裂缝区域进行细致的划分，分块拟合可以更大程度的提高裂缝面积的精度，减少计算误差，从而得到更精细的裂缝区域面积。
Pavement crack detection and small surface element fitting detection method based on YOLOv5
基于YOLOv5的路面裂缝检测及小面元拟合检测方法
WANG CHONGCHANG (Autor:in) / ZHENG SIWEN (Autor:in) / SUN SHANGYU (Autor:in) / JANG JIN-HYUK (Autor:in)
12.05.2023
Patent
Elektronische Ressource
Chinesisch
IPC:
G06T
Bilddatenverarbeitung oder Bilddatenerzeugung allgemein
,
IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
/
E01C
Bau von Straßen, Sportplätzen oder dgl., Decken dafür
,
CONSTRUCTION OF, OR SURFACES FOR, ROADS, SPORTS GROUNDS, OR THE LIKE
/
G06N
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
,
Rechnersysteme, basierend auf spezifischen Rechenmodellen
/
G06V
Tunnel lining crack detection model based on improved YOLOv5
| Elsevier | 2024
Tunnel lining crack detection model based on improved YOLOv5
| Elsevier | 2024