A NOVEL OBJECT DETECTION FRAMEWORK BASED ON INCREMENTAL LEARNING IN AN INDUSTRIAL SAFETY INSPECTION SYSTEM

Sun Jiawei,∗ Hao Jia,∗ Li Yimin,∗ and Yan Yan∗

References

  1. [1] Y. Bai, Y. Guo, Q. Zhang, B. Cao, and B. Zhang, Multi-network fusion algorithm with transfer learning for greencucumber segmentation and recognition under complex naturalenvironment, Computers and Electronics in Agriculture, 194,2022, 106789.
  2. [2] B. Bayram, T.B. Duman, and G. Ince, Real time detectionof acoustic anomalies in industrial processes using sequentialautoencoders, Expert Systems, 38(1), 2021, e12564.
  3. [3] B. Chen, K. Thandiackal, P. Pati, and O. Goksel, Generativeappearance replay for continual unsupervised domain adapta-tion, Medical Image Analysis, 89, 2023 102924.
  4. [4] M. Chen, L. Yu, C. Zhi, R. Sun, S. Zhu, Z. Gao, Z. Ke,M. Zhu, and Y. Zhang, Improved faster R-CNN for fabricdefect detection based on gabor filter with genetic algorithmoptimization, Computers in Industry, 134, 2022, 103551.
  5. [5] Y. Chen, S. Chen, Y. Deng, and K. Wang, HA-Transformer:Harmonious aggregation from local to global for objectdetection, Expert Systems with Applications, 230, 2023, 120539.
  6. [6] M. Cheng, C. Xu, J. Wang, W. Zhang, Y. Zhou, and J. Zhang,MicroCrack-Net: A deep neural network with outline profile-guided feature augmentation and attention-based multiscalefusion for microcrack detection of tantalum capacitors, IEEETransactions on Aerospace and Electronic Systems,58(6), 2022,5141–5152.
  7. [7] S. Dube, W.Y. Wan, and H. Nugroho, A novel approach ofIoT stream sampling and model update on the IoT edge devicefor class incremental learning in an edge-cloud system, IEEEAccess, 9, 2021, 29180–29199.
  8. [8] S. Ebrahimi, M. Elhoseiny, T. Darrell, and M. Rohrbach,Uncertainty-guided continual learning with Bayesian neuralnetworks, 2019, arXiv:1906.02425.
  9. [9] S. Ghazarian and M.A. Nematbakhsh, Enhancing memory-based collaborative filtering for group recommender systems,Expert Systems with Applications, 42(7), 2015, 3801–3812.
  10. [10] K. He, X. Zhang, S. Ren, and J. Sun, Deep residual learningfor image recognition, in Proceedings of the IEEE Conferenceon Computer Vision and Pattern Recognition, Las Vegas, NV,2016, 770–778.
  11. [11] X. He and H. Jaeger Overcoming catastrophic interferenceusing conceptor-aided backpropagation, in Proceeding of the6th International Conference on Learning Representations,Vancouver, BC, 2018, 1–11.
  12. [12] G. Hinton, Distilling the knowledge in a neural network, 2015,arXiv:1503.02531.
  13. [13] N. Hu, L. Ding, L. Men, W. Zhou, W. Zhang, and R. Yin,Dual visual inspection for automated quality detection andprinting optimization of two-photon polymerization based ondeep learning, Journal of Intelligent Manufacturing, 2024,1–13.
  14. [14] L.P. Jain, W.J. Scheirer, and T.E. Boult, Multi-class open setrecognition using probability of inclusion, in Proceeding of theECCV, Cham, 2014, 393–409.
  15. [15] A. Ji, J. Pang, and H. Qiu, Support vector machine forclassification based on fuzzy training data, Expert Systems withApplications, 37(4), 2010, 3495–3498.
  16. [16] L. Jiao, F. Zhang, F. Liu, S. Yang, L. Li, Z. Feng, and R. Qu, Asurvey of deep learning-based object detection, IEEE Access, 7,2019,128837–128868.
  17. [17] K.S. Kalyan, A. Rajasekharan, and S. Sangeetha, AMMU: Asurvey of transformer-based biomedical pretrained languagemodels, Journal of Biomedical Informatics, 126, 2022, 103982.
  18. [18] J. Kirkpatrick, R. Pascanu, N. Rabinowitz, J. Veness, G.Desjardins, A.A. Rusu, K. Milan, J. Quan, T. Ramalho, A.Grabska-Barwinska, and D. Hassabis, Overcoming catastrophicforgetting in neural networks, Proceedings of the NationalAcademy of Sciences, 114(13), 2017, 3521–3526.
  19. [19] Y. Kong, L. Liu, H. Chen, J. Kacprzyk, and D. Tao, Overcomingcatastrophic forgetting in continual learning by exploringeigenvalues of Hessian matrix, IEEE Transactions on NeuralNetworks and Learning Systems, 2023.
  20. [20] Y. Li and F. Li Growing deep echo state network with supervisedlearning for time series prediction, Applied Soft Computing,128, 2022, 109454.
  21. [21] Y. Li, Z. Wang, F. Li, Y. Li, X. Zhang, H. Shi, L. Dong, andW. Ren, An ensembled remaining useful life prediction methodwith data fusion and stage division, Reliability Engineering &System Safety, 242, 2024, 109804.
  22. [22] T. Lin, Focal loss for dense object detection, 2017,arXiv:1708.02002.
  23. [23] H. Liu, Y. Zhou, B. Liu, J. Zhao, R. Yao, and Z. Shao,Incremental learning with neural networks for computervision: A survey, Artificial Intelligence Review, 56(5), 2023,4557–4589.
  24. [24] Y. Lu, E.J. Ding, J. Du, G.C. Chen, and Y. Zheng, Safetydetection approach in industrial equipment based on RSSD withadaptive parameter optimization algorithm, Safety Science,125, 2020, 104605.
  25. [25] D. Miller, L. Nicholson, F. Dayoub, and N. S¨underhauf,Dropout sampling for robust object detection in open-setconditions, in Proceeding of the IEEE International Conferenceon Robotics and Automation (ICRA), Brisbane, QLD, 2018,3243–3249.
  26. [26] R. Mohandas, M. Southern, E. O’Connell, and M. Hayesm,A Survey of incremental deep learning for defect detectionin manufacturing, Big Data and Cognitive Computing, 8(1),2024, 7.
  27. [27] H. Qu, H. Rahmani, L. Xu, B. Williams, and J. Liu, Recentadvances of continual learning in computer vision: An overview,2021, arXiv:2109.11369.
  28. [28] J. Redmon and A. Farhadi, YOLO9000: better, faster, stronger,in Proceedings of the IEEE Conference on Computer Visionand Pattern Recognition, 2017, 7263–7271.
  29. [29] S. Ren, X. Pan, W. Zhao, B. Nie, and B. Han, Dynamicgraph transformer for 3D object detection, Knowledge-BasedSystems, 259, 2023, 110085.
  30. [30] Z. Ren, F. Fang, N. Yan, and Y. Wu, State of the art in defectdetection based on machine vision, International Journal ofPrecision Engineering and Manufacturing-Green Technology,9(2), 2022, 661–691.
  31. [31] W.J. Scheirer, A. de Rezende Rocha, A. Sapkota, and T.E.Boult, Toward open set recognition, IEEE Transactions onPattern Analysis and Machine Intelligence, 35(7), 2012,1757–1772.
  32. [32] B. Wang, P. Jiang, Z. Liu, Y. Li, J. Cao, and Y. Li,An adaptive lightweight small object detection method forincremental few-shot scenarios of unmanned surface vehicles,Engineering Applications of Artificial Intelligence, 133, 2024,107989.
  33. [33] P. Wang, H. Xiong, and H. He, Bearing fault diagnosis undervarious conditions using an incremental learning-based multi-task shared classifier, Knowledge-Based Systems, 266, 2023,110395.
  34. [34] Z. Wang, Y. Ta, W. Cai, and Y. Li, Research on aremaining useful life prediction method for degradationangle identification two-stage degradation process, MechanicalSystems and Signal Processing, 184, 2023, 109747.12
  35. [35] P. Viola and M. Jones, Rapid object detection using aboosted cascade of simple features, in Proceedings of the IEEEComputer Society Conference on Computer Vision and PatternRecognition CVPR, Kauai, HI, 2001, 1.
  36. [36] R. Girshick, J. Donahue, T. Darrell, and J. Malik, Richfeature hierarchies for accurate object detection and semanticsegmentation, Radioengineering, 85(9), 2021, 115–126.
  37. [37] S. Xing, Y. Lei, B. Yang, and N. Lu, Adaptive knowledgetransfer by continual weighted updating of filter kernels forfew-shot fault diagnosis of machines, IEEE Transactions onIndustrial Electronics, 69(2), 2021, 1968–1976.
  38. [38] N. Yang, Z. Wang, W. Cai, and Y. Li, Data regenerationbased on multiple degradation processes for remaining usefullife estimation, Reliability Engineering & System Safety, 229,2023, 108867.
  39. [39] S. Yin, S.X. Ding, X. Xie, and H. Luo, A review on basicdata-driven approaches for industrial process monitoring,IEEE Transactions on Industrial Electronics, 61(11), 2014,6418–6428.
  40. [40] Yuwono E I, Tjondonegoro D, Sorwar G, et al. Scalabilityof knowledge distillation in incremental deep learning for fastobject detection, Applied Soft Computing, 129, 2022, 109608.
  41. [41] F. Zenke, B. Poole, and S. Ganguli, Continual learning throughsynaptic intelligence, in Proceeding International Conferenceon Machine Learning, 2017, 3987–3995.
  42. [42] Q. Zhang, A new residual generation and evaluationmethod for detection and isolation of faults in non-linearsystems, International Journal of Adaptive Control and SignalProcessing, 14(7), 2000, 759–773.
  43. [43] T. Zhang, Z. Wang, F. Li, H. Zhong, X. Hu, W. Zhang, D.Zhang, and X. Liu, Automatic detection of surface defects basedon deep random chains, Expert Systems with Applications, 229,2023, 120472.
  44. [44] M. Zhao, C. Yue, and X. Liu, Research on milling chatteridentification of thin-walled parts based on incremental learningand multi-signal fusion, The International Journal of AdvancedManufacturing Technology, 125(9), 2023, 3925–3941.
  45. [45] A. Rodriguez, M. Sanchez, and T. Li, CBCL-PR: A cognitivelyinspired model for class-incremental learning in robotics,International Journal of Robotics and Automation, 40(3), 2024,215–230, DOI: http://dx.doi.org/10.2316/J.2024.403-012.
  46. [46] K. Wang, L. Zhao, and J. Gao, VIPeR: Visual incrementalplace recognition with adaptive mining and lifelong learning,International Journal of Robotics and Automation, 41(1), 2025,45–62, DOI: http://dx.doi.org/10.2316/J.2025.411-005.
  47. [47] W. Li, Y. Liu, J. Yang, and W. Wu, A new conjugategradient method with smoothing L1/2 regularization basedon a modified secant equation for training neural networks,Neural Processing Letters, 48(2), 2018, 955–978, DOI:http://dx.doi.org/10.1007/s11063-017-9737-9.
  48. [48] J. Yu, L. Ma, Z. Li, Y. Peng, and S. Xie, Open-world object detection via discriminative class prototypelearning, in Proceedings of the International Conference onImage Processing (ICIP), Bordeaux, 2022, 626–630, DOI:http://dx.doi.org/10.1109/ICIP46576.2022.9897461.

Important Links:

Go Back