Visual detection of sand concrete cracks in railway tunnel lining mechanism based on Mask R-CNN model
-
摘要:
铁路隧道衬砌结构裂缝图像具有复杂的灰度分布和变化特征,局部和全局的多特征信息会干扰跟踪方向和边界跟踪参数,模型可扩展性受限,检测准确率较低。为此,提出基于Mask R-CNN模型的铁路隧道衬砌机制砂混凝土裂缝视觉检测方法。首先输入分段线性变换后的砂混凝土裂缝图像,抽取阈值,生成连通域标识,再以像素点为背景点,在Mask R-CNN模型中,同时检测裂缝区域的位置和标记像素级的边缘掩膜,判定裂缝边界起点与裂缝宽度；然后进行累加视觉检测方法设计,按照裂缝的几何特征以及排序结果,求解裂缝长度,获得完整的裂缝轮廓。试验结果表明,所提方法可以较为完整地检测所有关键位置,裂缝参数信息检测准确率较高；迭代次数升高后,检测结果受到的影响较小,可扩展性得到了改善,可适应任务需求,具有较好的应用价值。
-
关键词:
- Mask R-CNN模型 /
- 铁路隧道 /
- 衬砌机制 /
- 砂混凝土裂缝 /
- 视觉检测
Abstract:The crack image of railway tunnel lining structure has complex grayscale distribution and variation characteristics, and local and global multi feature information interferes with tracking direction and boundary tracking parameters. The model’s scalability is limited, and the detection accuracy is low. Therefore, a visual detection of sand concrete cracks in railway tunnel lining mechanism based on Mask R-CNN model was proposed. Firstly, after inputting the segmented linear transformation of the sand concrete crack image, a threshold was extracted to generate a connected domain identifier. Pixel points were used as background points in the Mask R-CNN model to simultaneously detect the position of the crack area and mark pixel level edge masks, the starting point and width of the crack boundary was determined, and a cumulative visual detection method was designed. Based on the geometric characteristics and sorting results of the cracks, the crack length was calculated, and the complete crack contour was obtained. The experimental results showed that after using the method proposed in this paper, all key positions can be detected completely; After increasing the number of iterations, the detection results were less affected, indicating that its scalability had been improved and can adapt to task requirements; and therefore, it has good application value.
-
Keywords:
- Mask R-CNN model /
- railway tunnel /
- lining mechanism /
- sand concrete crack /
- visual inspection
-
-
[1] 王平让,黄宏伟,薛亚东 ．隧道衬砌裂缝机器视觉检测系统研制与应用[J]．公路,2022,67(10):439-446. [2] 毕东月 ．基于深度学习的输煤皮带故障视觉检测方法研究[J]．中国安全生产科学技术,2021,17(8):84-90. [3] 王保宪,白少雄,赵维刚 ．基于特征增强学习的路面裂缝病害视觉检测方法[J]．铁道科学与工程学报,2022,19(7):1927-1935. [4] 许霄煜,郭振武,周唯逸,等 ．基于目标提取与SVM的排洪隧洞裂缝视觉检测算法研究[J]．有色金属工程,2022,12(6):101-111. [5] 马晓雄,熊晓燕,兰媛,等 ．金属棒材表面缺陷的机器视觉检测方法研究[J]．机械设计与制造,2023(4):196-200,205. [6] 左旺,陈仲生,李潮林,等 ．基于正样本的产品表面缺陷视觉检测方法[J]．湖南工业大学学报,2023,37(1):69-74. [7] 郭井宽,张森,张延松 ．管屏拼焊焊缝表面缺陷的激光视觉检测方法[J]．焊接,2022(12):30-35. [8] 严宇,张宏伟,肖奕,等 ．基于轻量级卷积网络的电力绝缘子视觉检测[J]．电力科学与工程,2022,38(12):38-46. [9] 付学俊 ．视觉检测技术在带式输送机堆煤故障监测中的应用[J]．机械管理开发,2022,37(12):187-188,191. [10] 罗文彬,刘敏,李琳,等 ．基于视觉检测的红外与可见光图像区域匹配研究[J]．激光杂志,2023,44(2):186-190.
下载:
计量
- 文章访问数: 26
- HTML全文浏览量: 2
- PDF下载量: 11
