Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Road shoulder disease detection method for road maintenance
The invention belongs to the technical field of visual inspection, and particularly relates to a road shoulder disease detection method for road maintenance, which comprises the following steps of: self-making a road shoulder disease data set for road maintenance; labeling the collected images, and dividing data sets into a training set, a verification set and a test set according to a proportion; a multi-scale cavity attention MSDA module is introduced into the last layer of the neck of the original YOLOv8 model to capture multi-scale information; in the YOLOv8, a part of C2f modules are replaced by C2fGDC modules; and carrying out model training by using an improved YOLOv8n algorithm RM-YOLO, and applying the trained model to road shoulder disease detection. By fusing the above improved modules, better detection precision can be realized on the basis of not increasing parameter quantity and calculation cost.
本发明属于视觉检测技术领域，具体涉及一种用于道路养护的路肩病害检测方法，包括如下步骤：自制用于道路养护的路肩病害数据集；对采集图像进行标签标注，并按照比例进行数据集划分，包括训练集、验证集和测试集；在YOLOv8原模型的颈部最后一层引入多尺度空洞注意力MSDA模块，以捕捉多尺度信息；在YOLOv8中将部分C2f模块替换为C2fGDC模块；改进后的YOLOv8n算法RM‑YOLO进行模型训练，将训练后的模型应用于路肩病害检测。通过融合上述改进模块，在不增加参数量和计算成本基础上可以实现较好的检测精度。
Road shoulder disease detection method for road maintenance
The invention belongs to the technical field of visual inspection, and particularly relates to a road shoulder disease detection method for road maintenance, which comprises the following steps of: self-making a road shoulder disease data set for road maintenance; labeling the collected images, and dividing data sets into a training set, a verification set and a test set according to a proportion; a multi-scale cavity attention MSDA module is introduced into the last layer of the neck of the original YOLOv8 model to capture multi-scale information; in the YOLOv8, a part of C2f modules are replaced by C2fGDC modules; and carrying out model training by using an improved YOLOv8n algorithm RM-YOLO, and applying the trained model to road shoulder disease detection. By fusing the above improved modules, better detection precision can be realized on the basis of not increasing parameter quantity and calculation cost.
本发明属于视觉检测技术领域，具体涉及一种用于道路养护的路肩病害检测方法，包括如下步骤：自制用于道路养护的路肩病害数据集；对采集图像进行标签标注，并按照比例进行数据集划分，包括训练集、验证集和测试集；在YOLOv8原模型的颈部最后一层引入多尺度空洞注意力MSDA模块，以捕捉多尺度信息；在YOLOv8中将部分C2f模块替换为C2fGDC模块；改进后的YOLOv8n算法RM‑YOLO进行模型训练，将训练后的模型应用于路肩病害检测。通过融合上述改进模块，在不增加参数量和计算成本基础上可以实现较好的检测精度。
Road shoulder disease detection method for road maintenance
用于道路养护的路肩病害检测方法
MA XIAOXIONG (Autor:in) / ZHOU CONGLING (Autor:in) / WANG YONGQIANG (Autor:in) / TIAN ZHEN (Autor:in) / LI JIANJIA (Autor:in) / WANG JIAHAO (Autor:in)
23.08.2024
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