摘要
为更高效率地确定隧道围岩完整性程度,该文提出了一种基于轻量级神经网络MobileNet-v2的隧道围岩岩体完整性程度的精确确定方法。首先,将图像进行灰度化、图像降噪以及裂隙边缘检测处理;然后,利用MobileNet-v2轻量级神经网络模型在Image-Net数据集上进行预训练,并与迁移学习相结合完成训练集、验证集以及测试集的数据检测;最后,与传统神经网络RestNet-50、VGG16做对比试验。通过裂隙面积、宽度和长度的识别,引入裂隙比Ks作为评判围岩完整性程度的指标。结果表明:(1)在准确率、损失值和训练时间等方面,MobileNet-v2模型明显优于VGG16和RestNet-50模型;(2) MobileNet-v2模型的准确率最高,验证集准确率可达到94%左右;(3)与现场试验结果对比表明,使用数字图像处理方法评判岩体完整性具有较高的准确性和可行性。
To determine the integrity degree of tunnel surrounding rock more efficiently,a precise determination method for the integrity degree of tunnel surrounding rock based on the lightweight neural network MobileNet-v2 was proposed.Firstly,the image was grayscaled and denoised,and the edges of cracks were detected.Then,the lightweight neural network MobileNet-v2 model was pre-trained on the ImageNet dataset,and combined with transfer learning to complete data detection on the training,validation,and testing sets.Finally,a comparative experiment was conducted with traditional neural networks RestNet-50 and VGG16.By identifying the area,width,and length of cracks,the crack ratio Ks was introduced as an indicator to evaluate the integrity of the surrounding rock.The results show that:①in terms of accuracy,loss value,and training time,the MobileNet-v2 model is significantly better than the VGG16 and RestNet-50 models;②the MobileNet-v2 model has the highest accuracy,with a validation set accuracy of around 94%;③by comparing with the results of on-site experiments,it is proven that using digital image processing methods to evaluate the integrity of rock masses has high accuracy and feasibility.
作者
柳厚祥
杨彩霞
LIU Houxiang;YANG Caixia(School of Civil and Environmental Engineering,Changsha University of Science&Technology,Changsha,Hunan 410114,China)
出处
《中外公路》
2025年第6期227-233,共7页
Journal of China & Foreign Highway
基金
国家自然科学基金资助项目(编号:52078060)
湖南省自然科学基金资助项目(编号:2024JJ5036)。
