计算机科学 ›› 2024, Vol. 51 ›› Issue (11A): 231000082-6.doi: 10.11896/jsjkx.231000082
曹庆园, 朱建鸿
CAO Qingyuan, ZHU Jianhong
摘要： 为解决混凝土砂石骨料复杂种类识别精度低的问题,实现砂石骨料种类自动识别,提出了一种适用于混凝土砂石骨料种类识别的CM-ResNet18网络模型。首先采集骨料图像数据集,并采用数据增强方法增加样本以提升模型的鲁棒性;其次选择ResNet18模型作为主干网络,融合CBAM模块和MHSA模块,以增强模型的特征提取能力;最后增加Dropout函数以提高神经网络的泛化性能,同时在训练中引入迁移学习以加快网络收敛速度,并增大最后一层学习率,使其更好地适应训练数据并提高模型性能。实验结果表明,CM-ResNet18模型在原材料识别中取得了高达99.09%的准确率。与其他网络模型AlexNet,VGG19,EfficientNet,ResNet18,ResNet34相比,CM-ResNet18模型在识别准确率、精确率、召回率、F1-score上均有提高,表明该方法在混凝土砂石骨料识别中具有较高的实用性和可行性。
中图分类号:
| [1]FANG H Y,YANG J B,HUANG W J,et al.A kind of detection device and method for aggregate morphological quality:China,Fujian:CN106969708A[P].2017-07-21. [2]LI X.Study on aspect ratio detection method of concrete coarse aggregate based on image processing[J].Shanxi Architecture,2018,44(5):103-105. [3]LIU C,XU Q G,SHI B,et al.Digitalimage recognition method and application of rock particle and pore system[J].Chinese Journal of Geotechnical Engineering,2018,40(5):925-931. [4]XIONG C Y,WANG T.Image recognition of building compo-nents based on convolutional neural network[J].Computer Science,2021,48(S1):51-56. [5]CHENG G J,LI B,WAN X L,et al.Research on image classification of rock slices based on SqueezeNet convolutional neural network[J].Mineralogy and Petrology,2021,41(4):94-101. [6]YANG S R,SU M,WANG H.Sand and gravel aggregate classification model based on convolutional neural network[J].Sichuan Cement,2019,280(12):15-16. [7]YU L J,TONG X,SHEN G L,et al.Classification method ofsand and gravel aggregate using GA-SVM model[J].Journal of Huaqiao University(Natural Science Edition),2020,41(2):137-141. [8]YAN R,LIAO J D,WU X Y,et al.Research on sand and gravel aggregate classification method based on convolutional neural network[J].Progress in Laser & Optoelectronics,2021,58(20):211-218. [9]TUAN N M,KIM Y,LEE J Y,et al.Automatic stereo vision-based inspection system for particle shape analysis of coarse aggregates[J].Journal of Computing in Civil Engineering,2022,36(2):04021034. [10]HE K,ZHANG X,REN S,et al.Deep Residual Learning for Image Recognition[J].CoRR,2015,abs/1512.03385. [11]FENG K,CHASPARI T.A Review of Generalizable Transfer Learning in Automatic Emotion Recognition[J].Frontiers in Computer Science,2020,2:9. [12]HE K,ZHANG X,REN S,et al.Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:770-778. [13]WOO S,PARK J,LEE J Y,et al.Cbam:convolutional block attention module[C]//Proceedings of the European Conference on Computer Vision(ECCV).2018:3-19. [14]DOSOVITSKIY A,BEYER L,KOLESNIKOV A,et al.An image is worth 16x16 words:Transformers for image recognition at scale[J].arXiv:2010.11929,2020. [15]LYU C,HEYER P,GOOSSENS B,et al.An unsupervisedtransfer learning framework for visible-thermal pedestrian detection[J].Sensors,2022,22(12):4416. [16]KINGMA D,BA J.Adam:a method for stochastic optimization[C]//International Conference on Learning Representations(ICLR).2015:1-15. [17]DU Z W,ZHOU H,LI C Y,et al.Review of small target detection algorithms for deep convolutional neural networks[J].Computer Science,2022,49(12):205-218. [18]KRIZHEVSKY A,SUTSKEVER I,HINTON G E.Imagenetclassification with deep convolutional neural networks[C]//Proceedings of the 26th Conference on Neural Information Processing Systems(NIPS).Lake Tahoe,USA,2012:1106-1114. [19]SIMONYAN K,ZISSERMAN A.Very Deep Convolutional Networks for Large-Scale Image Recognition.[J].CoRR,2014,abs/1409.1556. [20]TAN M,LE Q.Efficientnet:Rethinking model scaling for convolutional neural networks[C]//International Conference on Machine Learning.PMLR,2019:6105-6114. [21]BANAN A,NASIRI A,TAHERI-GARAVAND A.Deep learning-based appearance features extraction for automated carp species identification[J].Aquacultural Engineering,2020,89:134-139. |
|
||
首页