改进YOLOv7的沥青路面病害检测

doi:10.3778/j.issn.1002-8331.2301-0098

摘要/Abstract

摘要： 针对传统卷积网络对沥青路面病害检测时存在的检测精度低、定位不准等问题，提出一种基于改进YOLOv7的沥青路面病害检测算法。针对激光图像的成像特征，使用组合滤波-三直方图均衡化算法弱化背景环境干扰；使用K-means++聚类算法进行初始锚框设置来加快模型收敛速度；然后将多头自注意力机制与最大池化层结合代替模型主干框架中部分卷积层，提高卷积网络对于目标物全局特征学习能力；使用漏斗激活函数F-ReLU作为主干网络中的激活函数以扩大卷积层的感受野范围；最后使用A-SIOU损失函数优化模型边界框回归，加快模型收敛的同时提高训练精度。实验结果表明，改进后的检测算法对病害检测的平均精度均值、精确率和召回率相较原网络提升了7.7、9.4与5.8个百分点，具有较好的识别精度。在实际工程应用中，对各路段的路面状况指数的计算偏差均小于1%，对推进路面病害的智能化检测具有重要意义。

关键词: 路面病害检测, 深度学习, 激光图像, 多头自注意力机制, 损失函数, 激活函数

Abstract: Aiming at the problems of low detection accuracy and inaccurate positioning in the detection of asphalt pavement diseases by traditional convolution network, an asphalt pavement disease detection algorithm based on improved YOLOv7 is proposed. Firstly, according to the imaging characteristics of the laser image, the combined filter-three histogram equalization algorithm is used to weaken the background environment interference. The K-means++clustering algorithm is used to set the initial anchor frame to speed up the convergence of the model. Then, the multi-head self-attention mechanism is combined with the maximum pooling layer to replace part of the convolution layer in the model backbone framework to improve the ability of the convolution network to learn the global features of the target. The funnel activation function F-ReLU is used as the activation function in the backbone network to expand the receptive field of the convolution layer. Finally, the A-SIOU loss function is used to optimize the model boundary box regression to accelerate the model convergence and improve the training accuracy. The experimental results show that the average precision, precision and recall rate of the improved detection algorithm for disease detection are 7.7, 9.4 and 5.8?percentage points higher than that of the original network, with better recognition accuracy. In practical engineering applications, the calculation deviation of the pavement condition index of each section is less than 1%, which is of great significance for promoting the intelligent detection of pavement diseases.

Key words: pavement diseases detection, deep learning, dimensional laser image, multi-head self-attention mechanism, loss function, activation function

倪昌双, 李林, 罗文婷, 秦勇, 杨振, 傅幼华. 改进YOLOv7的沥青路面病害检测[J]. 计算机工程与应用, 2023, 59(13): 305-316.

NI Changshuang, LI Lin, LUO Wenting, QIN Yong, YANG Zhen, FU Youhua. Disease Detection of Asphalt Pavement Based on Improved YOLOv7[J]. Computer Engineering and Applications, 2023, 59(13): 305-316.

参考文献

[1] BAI C L.Types and causes analysis of highway surface diseases[J].Science & Technology Vision，2015（1）：244-245.
[2] 侯越，张慧婷，高智伟，等.基于数据深度增强的路面病害智能检测方法研究及比较[J].北京工业大学学报，2022，48（6）：622-634.
HOU Y，ZHANG H T，GAO Z W，et al.Research and comparison of intelligent detection methods of pavement distress based on deep data augmentation[J].Journal of Beijing University of Technology，2022，48（6）：622-634.
[3] 朱平哲.路面裂缝图像检测算法研究[D].洛阳：河南科技大学，2012.
ZHU P Z.Research on algorithms of the pavement crack image detection[D].Luoyang：Henan University of Science and Technology，2012.
[4] 王青宁.基于图像处理的路面病害检测方法研究[D].南京：南京信息工程大学，2022.
WANG Q N.Research on pavement disease detection method based on image processing[D].Nanjing：Nanjing University of Information Science & Technology，2022.
[5] 邓砚学，张志华，张新秀.一种新的沥青路面灌封裂缝自动提取方法[J].科学技术与工程，2022，22（16）：6687-6694.
DENG Y X，ZHANG Z H，ZHANG X X.A novel approach for automatic extraction asphalt pavement sealed crack[J].Science Technology and Engineering，2022，22（16）：6687-6694.
[6] 石丽，裴莉莉，陈昊，等.改进RetinaNet的水泥路面露骨病害检测[J].计算机系统应用，2022，31（04）：352-359.
SHI L，FEI L L，CHEN H，et al.Detection of exposed distress of cement pavement based on improved RetinaNet[J].Computer Systems & Applications，2022，31（4）：352-359.
[7] SEO H.Deep learning-based thermal image analysis for pavement defect detection and classification considering complex pavement conditions[J].Remote Sensing，2021，14.
[8] 许慧青，陈斌，王敬飞，等.基于卷积神经网络的细长路面病害检测方法[J].计算机应用，2022，42（1）：265-272.
XU H Q，CHEN B，WANG J F，et al.Elongated pavement distress detection method based on convolutional neural network[J].Journal of Computer Applications，2022，42（1）：265-272.
[9] MANDAL V，LAN U，ADU-GYANFI Y.Automated road crack detection using deep convolutional neural networks[C]//Proceedings of the 2018 IEEE International Conference on Big Data，2018.
[10] 贾晨.基于YOLOv4的公路路面病害检测方法研究[D].南昌：华东交通大学，2021.
JIA C.Research on the detection method of road surface disease based on YOLOv4[D].Nanchang：East China Jiaotong University，2021.
[11] 罗晖，贾晨，李健.基于改进YOLOv4的公路路面病害检测算法[J].激光与光电子学进展，2021，58（14）：336-344.
LUO H，JIA C，LI J.Road surface disease detection algorithm based on improved YOLOv4[J].Laser & Optoelectronics Progress，2021，58（14）：336-344.
[12] WANG C Y，BOCHKOVSKIY A，LIAO H.YOLOv7：trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[J].arXiv：2207.02696，2022.
[13] 张正勇，桂冬冬，马蕴文，等.高维激光拉曼光谱的构建与降噪处理评价研究[J].应用激光，2018，38（3）：468-473.
ZHANG Z Y，GUI D D，MA Y W，et al.Evaluation on the construction and noise reduction of high-dimensional laser Raman spectroscopy[J].Applied Laser，2018，38（3）：468-473.
[14] 管继祥，李林，张超，等.基于组合滤波的M-MHSAOS图像固定模式噪声降噪算法[J].应用激光，2022，42（4）：173-180.
GUAN J X，LI L，ZHANG C，et al.Fixed pattern noise removal algorithm for M?MHSAOS image based on combines filter[J].Applied Laser，2022，42（4）：173-180.
[15] 常戬，贺春泽，董育理，等.改进双边滤波和阈值函数的图像增强算法[J].计算机工程与应用，2020，56（3）：207-213.
CHANG J，HE C Z，DONG Y L，et al.Improved image enhancement algorithm for bilateral filtering and threshold function[J].Computer Engineering and Applications，2020，56（3）：207-213.
[16] WANG C Y，BOCHKOVSKIY A，LIAO H Y M.Scaled-YOLOv4：scaling cross stage partial network[C]//Proceedings of the 2021 IEEE /CVF Conference on Computer Vision and Pattern Recognition.New York：IEEE Press，2021：13024-13033.
[17] DING X H，ZHANG X Y，MA N N，et al.RepVGG：making VGG-style ConvNets great again[C]//Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition.New York：IEEE Press，2021：13728-13737.
[18] RAMACHANDRAN P，ZOPH B，LE Q V.Searching for activation functions[J].arXiv：1710.05941，2017.
[19] MA N N，ZHANG X Y，SUN J.Funnel activation for visual recognition[C]//Proceedings of the European Conference on Computer Vision，2020：351-368.
[20] SRINIVAS A，LIN T Y，PARMAR N，et al.Bottleneck transformers for visual recognition[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition，2021：16519-16529.
[21] GEVORGYAN Z.SIoU loss：more powerful learning for bounding box regression[J].arXiv：2205.12740，2022.
[22] HE J，ERFANI S，MA X，et al.Alpha-IOU：a family of power intersection over union losses for bounding box regression[C]//Advances in Neural Information Processing Systems，2021：20230-23242.