浏览全部资源
扫码关注微信
收稿日期：2024-11-27，
修回日期：2025-03-06，
录用日期：2025-03-10，
网络出版日期：2025-03-12，
移动端阅览
目的
2
路面缺陷检测是道路维护和管理的重要环节，人工智能技术能够极大地提升路面缺陷检测的性能和效率，为解决当前路面缺陷检测算法难以满足在计算资源受限设备上进行实时检测的问题，本文基于YOLOv8（you only look once version 8）目标检测框架，结合部分卷积（partial convolution）与初始深度卷积（inception depthwise convolution）机制提出了一种轻量级的路面缺陷检测算法YOLOv8n-PIVI。
方法
2
该算法在骨干特征提取网络引入PartialBlock降低模型参数量，同时引入IDBlock（inception depthwise block）丰富网络的特征提取能力并进一步降低计算复杂度；在特征融合网络引入VanillaBlock减少网络参数的同时优化网络特征融合能力；引入ID-Detect（inception depthwise detect）检测头，极大降低模型的计算复杂度并优化检测头对于不同尺度特征的解析能力。
结果
2
在Pothole Dataset数据集上的实验结果显示，本文提出的算法mAP50（mean average precision at 50% intersection over union）达到了0.55，较基线算法提升了3.5个百分点，FPS（frames per second）达到了243，较基线算法提升了43，并且模型参数量和计算量仅为基线67%和72%，模型参数文件由6.3MB（mega byte）降低至4.3MB，内存占用较原模型降低了近1/3。在与目标检测领域较经典算法的对比实验中，本文所提出的算法在同参数量级模型中检测精度、计算复杂度及FPS较其他算法都有一定优势。此外，在RDD2022（road damage detector 2022）-China-MotorBike数据集和VOC2012（visual object classes 2012）数据集上的实验结果表明，本文所提算法具有良好的适应性和鲁棒性，能够适应不同的目标检测任务。
结论
2
综合实验结果表明，本文提出的算法能够以轻量级的计算资源需求，实现更高的目标检测性能，模型参数文件大小低至4.3MB，并且具有良好的适应性和鲁棒性，有助于降低路面缺陷检测算法的使用限制，拓宽路面缺陷检测系统的适用场景。
Objective
2
With the acceleration of global urbanization and the rapid development of intelligent transportation， road construction and maintenance have become an important part of urban management. With the rapid growth of urban population and the increase of motor vehicles， road facilities are under great pressure. Road surface defects， such as cracks and potholes， are prone to occur under the action of multiple factors such as frequent vehicle crushing and climate change. These defects not only affect driving comfort， but also potentially pose serious traffic safety risks. Pavement defect detection is an important part of road maintenance and management， which is related to traffic safety and road service life. Traditional pavement defect detection methods have disadvantages such as low efficiency， poor security， strong subjectivity and insufficient accuracy. With the development of computer technology and artificial intelligence， researchers have gradually applied machine vision and deep learning technology to pavement defect detection， which has greatly improved the performance and efficiency of pavement defect detection. However， due to the large amount of parameters and calculation， the traditional deep learning model is difficult to meet the real-time detection requirements on embedded devices and edge devices with limited hardware resources， which limits the promotion and application of deep learning technology in the field of road defect detection. In order to reduce the computational resource requirements and optimize the target detection performance of the pavement defect detection algorithm. Based on the YOLOv8 （you only look once version 8） object detection framework， combined with partial convolution and inception depthwise convolution mechanism， a lightweight pavement defect detection algorithm YOLOv8n-PIVI is proposed.
Method
2
Firstly， in order to reduce the number of parameters of the backbone network and improve the processing ability of the network for features of different scales， the PartialBlock module and IDBlock （inception depthwise block） module were introduced into the backbone network. While introducing the lightweight advantage of partial convolution， the PartialBlock module also supplements the 1 × 1 convolution to extract the overall information and arrange features， which not only makes up for the lack of feature analysis ability of partial convolution to a certain extent， but also helps partial convolution learn a more effective spatial feature extraction method. The IDBlock module inherits the advantages of partial lightweight and deeper feature analysis of inception depthwise convolution， and performs a more comprehensive analysis of features at a smaller cost through the traditional 1 × 1 convolution. In addition， both PartialBlock module and IDBlock module contain a MaxPool module in the tail to expand the receptive field of network features and reduce the number of network parameters. The organic combination of the PartialBlock module and the IDBlock module helps to reduce the number of backbone network parameters and optimize the feature extraction ability of the backbone network. Secondly， inspired by the excellent feature parsing ability and the applicability in resource-constrained scenarios of VanillaNet， we introduce VanillaBlock， the core module of VanillaNet， into the feature fusion network of YOLOv8n to optimize the feature fusion ability of the model and reduce the computational burden of the model. Finally， in order to reduce the computational cost of the detection head and improve its parsing ability of different scale features. In this paper， the inception depthwise convolution module with low overhead and multi-scale feature parsing ability is introduced to replace the two series 3 × 3 convolutions in the original detection head by combining the 1 × 1 convolution with low overhead. A more lightweight and efficient ID-Detect （inception depthwise detect） was proposed to make the detection head lighter and further improve the performance of the detection head.
Result
2
The comparison and ablation experiments are conducted on the Pothole Dataset， which consists of 1784 still images taken by vehicle-mounted cameras and other mobile devices. The dataset is divided into a training set （1265 images）， a validation set （401 images）， and a test set （118 images） with an 11 ： 3 ： 1 ratio. In order to verify the effectiveness of the improved model in optimizing the pavement defect detection effect， the improved model is compared with some classical algorithms in the field of object detection， including Centernet， RetinaNet， Faster R-CNN （faster region-based convolutional neural networks）， SSD （single shot multibox detector）， Efficientdet， YOLOv8. The experimental results show that the Precision and Recall of the optimization algorithm proposed in this paper are improved to a certain extent compared with the baseline， and the mAP50 （mean average precision at 50% intersection over union） reaches 0.55. It is 3.5 percentage points higher than the baseline algorithm's 0.515， the FPS （frames per second） reaches 243， whic
h is 43 higher than the baseline algorithm， and the parameter number of the model is reduced from 3.01 × 10
6
to 2.01 × 10
6
， which is only 67% of the baseline. The calculation amount is reduced from 8.1GFLOPs （giga floating point of operations） to 5.9GFLOPs， which is only 72% of the baseline. The model parameter file is reduced from 6.3MB （mega byte） to 4.3MB， and the memory occupation is reduced by nearly 1/3 compared with the original model. Compared with the classical algorithms in the field of object detection， the proposed algorithm has certain advantages in detection accuracy， computational complexity and FPS compared with other algorithms in the model of the same parameter magnitude. In the ablation experiment， the PartialBlock module， IDBlock module， VanillaBlock module and ID-Detect module introduced in this paper are individually added to the baseline model and removed from the final model. The effectiveness of the modules introduced in this paper to improve the performance of the baseline algorithm for pavement defect detection is verified. In the subsequent comparative analysis， the improvement of the optimization algorithm proposed in this paper compared with the baseline algorithm in pavement defect detection performance is shown in a more detailed and intuitive way through the comparison of experimental result index parameters， the comparison of training process index changes， and the comparison of pavement pothole detection effects. Subsequently， we also designed experiments to roughly simulate the performance of the pavement defect detection algorithm proposed in this paper in the actual situation， and analyzed the shortcomings of the algorithm. In addition， we also compare the proposed algorithm and the baseline algorithm on the RDD2022 （road damage detector 2022）-China-MotorBike dataset and VOC2012 （visual object classes 2012） dataset to analyze the robustness of the algorithm. The experimental results show that the proposed algorithm can adapt to the multi-type road defect
detection tasks in China on the RDD2022-China-MotorBike dataset and the multi-class daily object detection tasks on the VOC2012 dataset， while shows better object detection performance than the baseline model. This shows that the proposed algorithm can adapt to different object detection tasks， has good adaptability and robustness， and is expected to be applicable to more complex and diverse road defect detection scenarios.
Conclusion
2
The comprehensive experimental results show that the optimized algorithm YOLOv8n-PIVI proposed in this paper can achieve higher object detection performance with lightweight computing resource requirements， the model parameter file size is as low as 4.3MB， and the algorithm has good adaptability and robustness， which is helpful to reduce the use limit of road defect detection algorithms and broaden the applicable scene of pavement defect detection system.
Arya D ， Maeda H ， Ghosh S K ， Toshniwal D and Sekimoto Y . 2024 . RDD2022： A multi-national image dataset for automatic Road Damage Detection . Geoscience Data Journal ， 11 （ 4 ）： 846 - 862 ［ DOI： 10.48550/arXiv.2209.08538 http://dx.doi.org/10.48550/arXiv.2209.08538 ］
Boureau Y L ， Ponce J and LeCun Y . 2010 . A theoretical analysis of feature pooling in visual recognition //Proceedings of the 27th International Conference on Machine Learning （ICML-10）： 111 - 118
Cai Z W and Casconcelos N . 2018 . Cascade R-CNN： Delving into High Quality Object Detection/ /Proceedings of the IEEE confere nce on computer vision and pattern recognition： 6154 - 6162 ［ DOI： 10.1109/cvpr.2018.00644 http://dx.doi.org/10.1109/cvpr.2018.00644 ］
Carion N ， Massa F ， Synnaeve G ， Usunier N ， Kirillov A and Zagoruyko S . 2020 . End-to-end object detection with transformers // European conference on computer vision . Cham ： Springer International Publishing： 213 - 229 ［ DOI： 10.1007/978-3-030-58452-8_13 http://dx.doi.org/10.1007/978-3-030-58452-8_13 ］
Cheng K W and Zhang L X . 2023 . A Survey of Object Detection Models Based on Deep Learning . Computer Science and Technology . 2023， 2 （ 2 ）： 93 ［ DOI： 10.57237/j.cst.2023.02.006 http://dx.doi.org/10.57237/j.cst.2023.02.006 ］
Chen H T ， Wang Y H ， Guo J Y and Tao D C . 2023 . VanillaNet： the Power of Minimalism in Deep Learning ［EB/OL］.［ 2023-05-23 ］. https://arxiv.org/pdf/2305.12972 https://arxiv.org/pdf/2305.12972
Chen J R ， Kao S H ， He H ， Zhuo W P ， Lee C and Chan S H . 2023 . Run， Don't Walk： Chasing Higher FLOPS for Faster Neural Networks/ /2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition： 12021 - 12031 ［ DOI： 10.1109/CVPR52729.2023.01157 http://dx.doi.org/10.1109/CVPR52729.2023.01157 ］
Chen J Y ， Zhou C L ， Wang S H ， Xia L and Chen Z . 2023 . Research on the rapid detection method of road defects based on improved YOLOv5 . Electronic measurement technology ， 46 （ 10 ）： 129 - 135
陈建瑜 ， 邹春龙 ， 王生怀 ， 夏力 ， 陈哲 . 2023 . 改进YOLOv5的路面缺陷快速检测方法研究 . 电子测量技术 ， 46 （ 10 ）： 129 - 135 ［ DOI： 10.19651/j.cnki.emt.2211195 http://dx.doi.org/10.19651/j.cnki.emt.2211195 ］
Cong X H ， Li S X ， Chen F K ， Liu C and Meng Y . 2023 . A Review of YOLO Object Detection Algorithms based on Deep Learning . Frontiers in Computing and Intelligent Systems ， 4 （ 2 ）： 17 - 20 ［ DOI： 10.54097/fcis.v4i2.9730 http://dx.doi.org/10.54097/fcis.v4i2.9730 ］
Dai J F ， Li Y ， He K M and Sun J . 2016 . R-FCN： Object Detection via Region-based Fully Convolutional Networks ［EB/OL］.［ 2023-12-11 ］. https://arxiv.org/pdf/1605.06409 https://arxiv.org/pdf/1605.06409
Duan K W ， Bai S ， Xie L X ， Qi H G ， Huang Q M and Tian Q . 2019 . CenterNet： Keypoint Triplets for Object Detection/ /2019 IEEE/CVF International Conference on Computer Vision （ICCV）： 6569 - 6578 ［ DOI： 10.1109/iccv.2019.00667 http://dx.doi.org/10.1109/iccv.2019.00667 ］
Du F J ， Jiao S J . 2022 . Improvement of Lightweight Convolutional Neural Network Model Based on YOLO Algorithm and Its Research in Pavement Defect Detection . Sensors ， 22 （ 9 ）： 3537 ［ DOI： 10.3390/s22093537 http://dx.doi.org/10.3390/s22093537 ］
Girshick R ， Donahue J ， Darrell T and Malik J . 2014 . Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation/ /Proceedings of the IEEE conference on computer vision and pattern recognition： 580 - 587 ［ DOI： 10.1109/cvpr.2014.81 http://dx.doi.org/10.1109/cvpr.2014.81 ］
Girshick R . 2015 . Fast R-CNN// 2015 IEEE International Conference on Computer Vision （ICCV） ： 1440 - 1448 ［ DOI： 10.1109/iccv.2015.169 http://dx.doi.org/10.1109/iccv.2015.169 ］
Gopalakrishnan K . 2018 . Deep Learning in Data-Driven Pavement Image Analysis and Automated Distress Detection： A Review . Data ， 3 （ 3 ）： 28 ［ DOI： 10.3390/data3030028 http://dx.doi.org/10.3390/data3030028 ］
Gu Z Y ， Yu Y C and Zhe T T . 2023 . Rapider-YOLOX： An Efficient Lightweight Object Detection Network . Journal of Intelligent Science and Technology ， 5 （ 01 ）： 92 - 103
顾宙瑜 ， 於跃成 ， 者甜甜 . 2023 . Rapider-YOLOX：高效的轻量级目标检测网络 . 智能科学与技术学报 ， 5 （ 01 ）： 92 - 103 ［ DOI： 10.11959/j.issn.2096-6652.202303 http://dx.doi.org/10.11959/j.issn.2096-6652.202303 ］
He K M ， Gkioxari G ， Dollar P and Girshick R . 2020 . Mask R-CNN// Proceedings of the IEEE international conference on computer vision ： 2961 - 2969 ［ DOI： 10.1109/tpami.2018.2844175 http://dx.doi.org/10.1109/tpami.2018.2844175 ］
He K M ， Xiang Y Z ， Shao Q R and Jian S . 2016 . Deep residual learning for image recognition/ /Proceedings of the IEEE conference on computer vision and pattern recognition： 770 - 778 ［ DOI： 10.1109/cvpr.2016.90 http://dx.doi.org/10.1109/cvpr.2016.90 ］
Lai J ， Peng R H ， Sun D X and Huang J . 2024 . Camouflage object detection combining attention mechanism and multi-detection layer structure ， 29 （ 1 ）： 134 - 146
赖杰 ， 彭锐晖 ， 孙殿星 ， 黄杰 . 2024 . 融合注意力机制与多检测层结构的伪装目标检测 . 中国图象图形学报 ， 29 （ 1 ）： 134 - 146 ［ DOI： 10.11834/jig.221189 http://dx.doi.org/10.11834/jig.221189 ］
Li Y ， Bian S ， Wang C T and Lu W . 2023 . Efficient Deepfake Detection using CNN combined with Transformer ， 28 （ 3 ）： 804 - 819
李颖 ， 边山 ， 王春桃 ， 卢伟 . 2023 . CNN结合Transformer的深度伪造高效检测 . 中国图象图形学报 ， 28 （ 3 ）： 804 - 819 ［ DOI： 10.11834/jig.220519 http://dx.doi.org/10.11834/jig.220519 ］
Li Z H ， An J T ， Jia H Y and F Y . 2023 . Lightweight remote sensing image detection model combining rotation box and attention mechanism ， 28 （ 9 ）： 2706 - 2718
李朝辉 ， 安金堂 ， 贾红雨 ， 方艳 . 2023 . 结合旋转框和注意力机制的轻量遥感图像检测模型 . 中国图象图形学报 ， 28 （ 9 ）： 2706 - 2718 ［ DOI： 10.11834/jig.220839 http://dx.doi.org/10.11834/jig.220839 ］
Liang F T ， Yan Z ， Xin C ， Feng Z L ， Chen H Z and Xin J W . 2021 . Review of target detection technology based on deep learning //Proceedings of the 5th International Conference on Control Engineering and Artificial Intelligence： 132 - 135 ［ DOI： 10.1145/3448218.3448234 http://dx.doi.org/10.1145/3448218.3448234 ］
Liang L M ， Long P W ， Lu B H ， Ou Y Y ， Zeng L . 2024 . EHH-YOLOv 8 s： A Lightweight Steel Strip Surface Defect Detection Algorithm. Journal of Beijing University of Aeronautics and Astronautics ： 1 - 15
梁礼明 ， 龙鹏威 ， 卢宝贺 ， 欧燕燕 ， 曾璐 . 2024 . EHH-YOLOv 8 s：一种轻量级的带钢表面缺陷检测算法. 北京航空航天大学学报 ： 1 - 15 ［ DOI： 10.13700/j.bh.1001-5965.2024.0426 http://dx.doi.org/10.13700/j.bh.1001-5965.2024.0426 ］
Lin T Y ， Goyal P ， Girshick R ， He K M and Dollar P . 2017 . Focal Loss for Dense Object Detection/ /2017 IEEE International Conference on Computer Vision （ICCV）： 2999 - 3007 ［ DOI： 10.1109/iccv.2017.324 http://dx.doi.org/10.1109/iccv.2017.324 ］
Liu W ， Anguelov D ， Erhan D ， Szegedy C ， Reed S ， Fu C Y and Berg A C . 2016 . SSD： Single Shot MultiBox Detector // Computer Vision–ECCV 2016： 14th European Conference ， Amsterdam， The Netherlands， October 11 – 14 ， 2016， Proceedings， Part I 14. Springer International Publishing ： 21 - 37 ［ DOI： 10.1007/978-3-319-46448-0_2 http://dx.doi.org/10.1007/978-3-319-46448-0_2 ］
Liu Z ， Mao H Z ， Wu C Y ， Feichtenhofer C ， Darrell T and Xie S N . 2022 . A convnet for the 2020s/ /Proceedings of the IEEE/CVF conference on computer vision and pattern recognition： 11976 - 11986 ［ DOI： 10.1109/cvpr52688.2022.01167 http://dx.doi.org/10.1109/cvpr52688.2022.01167 ］
Nienaber S ， Booysen M and Kroon R . 2015 . Detecting Potholes Using Simple Image Processing Techniques and Real-World Footage// Cognitive Informatics and Soft Computing ： 893 - 902 ［ DOI： 10.13140/RG.2.1.3121.8408 http://dx.doi.org/10.13140/RG.2.1.3121.8408 ］
Nienaber S ， Kroon R and Booysen M . 2015 . A Comparison of Low-Cost Monocular Vision Techniques for Pothole Distance Estimation // 2015 IEEE symposium series on computational Intelligence . IEEE ： 419 - 426 ［ DOI： 10.1109/SSCI.2015.69 http://dx.doi.org/10.1109/SSCI.2015.69 ］
Qin R B ， Chen N J and Huang Y H . 2022 . EDDNet： An Efficient and Accurate Defect Detection Network for the Industrial Edge Environment/ /Proceedings of the 2022 lEEE 22nd Inter-national Conference on Software Quality， Reliability and Security（QRS）： 854 - 863 ［ DOI： 10.1109/QRS57517.2022.00090 http://dx.doi.org/10.1109/QRS57517.2022.00090 ］
Redmon J ， Divvala S ， Girshick R and Farhadi A . 2016 . You Only Look Once： Unified， Real-Time Object Detection/ /Proceedings of the IEEE conference on computer vision and pattern recognition （CVPR）： 799 - 788 ［ DOI： 10.1109/cvpr.2016.91 http://dx.doi.org/10.1109/cvpr.2016.91 ］
Ren S Q ， He K M ， Girshick R and Sun J . 2017 . Faster R-CNN： Towards Real-Time Object Detection with Region Proposal Networks . IEEE Transactions on Pattern Analysis and Machine Intelligence ， 39 （ 6 ）： 1137 - 1149 ［ DOI： 10.1109/tpami.2016.2577031 http://dx.doi.org/10.1109/tpami.2016.2577031 ］
Spencer B F ， Hoskere V and Narazaki Y . 2019 . Advances in Computer Vision-Based Civil Infrastructure Inspection and Monitoring . Engineering ， 5 （ 2 ）： 199 - 222 ［ DOI： 10.1016/j.eng.2018.11.030 http://dx.doi.org/10.1016/j.eng.2018.11.030 ］
Szegedy C ， Wei L ， Yang Q J ， Sermanet P ， Reed S ， Anguelov D ， Erhan D ， Vanhoucke V and Rabinovich A . 2015 . Going deeper with convolutions/ /Proceedings of the IEEE conference on computer vision and pattern recognition： 1 - 9 ［ DOI： 10.1109/cvpr.2015.7298594 http://dx.doi.org/10.1109/cvpr.2015.7298594 ］
Szegedy C ， Vanhoucke V ， Ioffe S ， Shlens J and Wojna Z . 2016 . Rethinking the Inception Architecture for Computer Vision/ /Proceedings of the IEEE conference on computer vision and pattern recognition： 2818 - 2826 ［ DOI： 10.1109/cvpr.2016.308 http://dx.doi.org/10.1109/cvpr.2016.308 ］
Tan M X ， Pang R M and Le Q V . 2020 . EfficientDet： Scalable and Efficient Object Detection/ /Proceedings of the IEEE/CVF conference on computer vision and pattern recognition： 10781 - 10790 ［ DOI： 10.1109/cvpr42600.2020.01079 http://dx.doi.org/10.1109/cvpr42600.2020.01079 ］
Terven J ， Cordova-Esparza D M and Romero-Gonzalez J A . 2023 . A comprehensive review of yolo architectures in computer vision： From yolov1 to yolov8 and yolo-nas . Machine Learning and Knowledge Extraction ， 5 （ 4 ）： 1680 - 1716 ［ DOI： 10.3390/make5040083 http://dx.doi.org/10.3390/make5040083 ］
Wang C M and Liu H . 2024 . YOLOv8-VSC： A Lightweight Steel Strip Surface Defect Detection Algorithm . Computer Science and Exploration ， 18 （ 01 ）： 151 - 160
王春梅 ， 刘欢 . 2024 . YOLOv8-VSC：一种轻量级的带钢表面缺陷检测算法 . 计算机科学与探索 ， 18 （ 01 ）： 151 - 160 ［ DOI： 10.3778/j.issn.1673-9418.2308060 http://dx.doi.org/10.3778/j.issn.1673-9418.2308060 ］
Wang H X ， Zhang Y S ， Song B ， Chen D S and Yang Y . 2023 . NAS-FPNLite object detection method Fusing cross-stage connection and inverted residual ， 28 （ 4 ）： 1004 - 1018
王红霞 ， 张永善 ， 宋邦 ， 陈德山 ， 杨益 . 2023 . 融合跨阶段连接与倒残差的NAS-FPNLite目标检测方法 . 中国图象图形学报 ， 28 （ 4 ）： 1004 - 1018 ［ DOI： 10.11834/jig.211099 http://dx.doi.org/10.11834/jig.211099 ］
Wang J ， Xu P Q ， Li L J and Zhang F . 2023 . DAssd-Net： A Lightweight Steel Surface Defect Detection Model Based on Multi-Branch Dilated Convolution Aggregation and Multi-Domain Perception Detection Head . Sensors ， 23 （ 12 ）： 5488 ［ DOI： 10.3390/s23125488 http://dx.doi.org/10.3390/s23125488 ］
Wan F ， Chen S ， Hong Y H ， Guang B L ， LI X and Teng X . 2022 . YOLO-LRDD： a lightweight method for road damage detection based on improved YOLOv5s . EURASIP Journal on Advances in Signal Processing ， 2022 （ 1 ）： 98 ［ DOI： 10.1186/s13634-022-00931-x http://dx.doi.org/10.1186/s13634-022-00931-x ］
Wei F ， Zhou J P ， Tan X ， Lin J ， Tian L and Wang H . 2024 . Lightweight YOLOv5 detection algorithm for low-altitude micro UAV . Photoelectron · Laser ， 35 （ 06 ）： 641 - 649
魏峰 ， 周建平 ， 谭翔 ， 林静 ， 田莉 ， 王虎 . 2024 . 针对低空微小型无人机的轻量型YOLOv5检测算法 . 光电子·激光 ， 35 （ 06 ）： 641 - 649 ［ DOI： 10.16136/j.joel.2024.06.0741 http://dx.doi.org/10.16136/j.joel.2024.06.0741 ］
Xiao Z J ， Lin B H and Qu H C . 2024 . Fusing Multiple mechanisms for SAR Ship Detection ， 29 （ 2 ）： 545 - 558
肖振久 ， 林渤翰 ， 曲海成 . 2024 . 融合多重机制的SAR舰船检测 . 中国图象图形学报 ， 29 （ 2 ）： 545 - 558 ［ DOI： 10.11834/jig.230166 http://dx.doi.org/10.11834/jig.230166 ］
Xu T F ， Huang H ， Zhang H M and Niu XF . 2024 . Lightweight road disease detection method Based on improved YOLOv8 . Computer Engineering and Application ， 60 （ 14 ）： 175 - 186
胥铁峰 ， 黄河 ， 张红民 ， 牛晓富 . 2024 . 基于改进YOLOv8的轻量化道路病害检测方法 . 计算机工程与应用 ， 60 （ 14 ）： 175 - 186 ［ DOI： 10.3778/j.issn.1002-8331.2402-0243 http://dx.doi.org/10.3778/j.issn.1002-8331.2402-0243 ］
Yu W H ， Zhou P ， Yan S C and Wang X C . 2024 . InceptionNeXt： When Inception Meets ConvNeXt/ /2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）： 5672 - 5683 ［ DOI： 10.1109/CVPR52733.2024.00542 http://dx.doi.org/10.1109/CVPR52733.2024.00542 ］
Zakeri H ， Nejad F M and Fahimifar A . 2017 . Image Based Techniques for Crack Detection，Classification and Quantification in Asphalt Pavement： A Review . Archives of Computational Methods in Engineering ， 24 ： 935 - 977 ［ DOI： 10.1007/s11831-016-9194-z http://dx.doi.org/10.1007/s11831-016-9194-z ］
Zeng Y F ， Y X ， Hua F ， Wang P P and Gu M . 2023 . Inwinding decoupled Lightweight Network for Panoramic wisdom tooth Detection ， 28 （ 8 ）： 2491 - 2504
曾怡峰 ， 姚潇 ， 华飞 ， 王佩佩 ， 顾敏 . 2023 . 面向全景智齿检测的内卷解耦轻量化网络 . 中国图象图形学报 ， 28 （ 8 ）： 2491 - 2504 ［ DOI： 10.11834/jig.220377 http://dx.doi.org/10.11834/jig.220377 ］
Zhang J ， Wan G ， Jiang M ， Lu G ， Tao X and Huang Z . 2023 . Small object detection in UAV image based on improved YOLOv5 . Systems Science & Control Engineering ， 11 （ 1 ）： 2247082 ［ DOI： 10.1080/21642583.2023.2247082 http://dx.doi.org/10.1080/21642583.2023.2247082 ］
Zhang D P ， He S J ， Wei Y Y ， Xu Y C ， Hu H M and Huang W J . 2024 . Lightweight road defect detection method based on dynamic deformable convolution . Journal of Computer Aided Design and Graphics ： 1 - 12
章东平 ， 何数技 ， 魏杨悦 ， 徐云超 ， 胡海苗 ， 黄文君 . 2024 . 基于动态可变形卷积的轻量化道路缺陷检测方法 . 计算机辅助设计与图形学学报 ， 1 - 12 ［ DOI： 10.3724/SP.J.1089.2024-00332 http://dx.doi.org/10.3724/SP.J.1089.2024-00332 ］
Zhou X Y ， Wang D Q and Krähenbühl P . 2019 . Objects as Points ［EB/OL］.［ 2019-04-25 ］. https://arxiv.org/pdf/1904.07850 https://arxiv.org/pdf/1904.07850
Zhou T ， Ye X Y ， Zhao Y N ， Lu H L and Liu F Z . 2024 . Cross-modal attention YOLOv5 for PET/CT lung tumor detection ， 29 （ 4 ）： 1070 - 1084
周涛 ， 叶鑫宇 ， 赵雅楠 ， 陆惠玲 ， 刘凤珍 . 2024 . 跨模态注意力YOLOv5的PET/CT肺部肿瘤检测 . 中国图象图形学报 ， 29 （ 4 ）： 1070 - 1084 ［ DOI： 10.11834/jig.230169 http://dx.doi.org/10.11834/jig.230169 ］
Zhuo W Q ， Li D Z ， Wang W and Dong J . 2023 . Data-free Model Compression for Lightweight Deepfake Detection ， 28 （ 3 ）： 820 - 835
卓文琦 ， 李东泽 ， 王伟 ， 董晶 . 2023 . 面向轻量级深度伪造检测的无数据模型压缩 . 中国图象图形学报 ， 28 （ 3 ）： 820 - 835 ［ DOI： 10.11834/jig.220559 http://dx.doi.org/10.11834/jig.220559 ］
相关作者
相关机构
京公网安备11010802024621