Page 100 - 《应用声学》2025年第1期

96 2025 年 1 月

IEEE Transactions on Ultrasonics Ferroelectrics and Fre- [24] Liu S, Qi L, Qin H, et al. Path aggregation network
quency Control, 2021, 68(10): 3126–3134. for instance segmentation[C]. In Proceedings of the 2018
[21] Gantala T, Balasubramaniam K. Automated defect recog- IEEE/CVF Conference on Computer Vision and Pattern
nition for welds using simulation assisted tfm imaging Recognition (CVPR),Salt Lake City, UT, USA, 18–23
with artificial intelligence[J]. Journal of Nondestructive June 2018, 2018.
Evaluation, 2021, 40(1): 28. [25] Zheng Z, Wang P, Liu W, et al. Distance-IoU loss:
[22] Wang C Y, Liao H, Wu Y H, et al. CSPNet: A new Faster and better learning for bounding box regres-
backbone that can enhance learning capability of CNN[C]. sion[C]//Proceedings of the AAAI Conference on Artifi-
In Proceedings of the 2020 IEEE/CVF Conference on cial Intelligence, 2020: 12993–13000.
Computer Vision and Pattern Recognition Workshops [26] Hu J, Shen L, Albanie S, et al. Squeeze-and-excitation
(CVPRW), Seattle, WA, USA, 14–19 June 2020, 2020. networks[J]. IEEE Transactions on Pattern Analysis and
[23] Lin T Y, Dollar P, Girshick R, et al. Feature pyra- Machine Intelligence, 2020, 42(8): 2011–2023.
mid networks for object detection[C]. In Proceedings of [27] Han G, Li T, Li Q, et al. Improved algorithm for insula-
the 2017 IEEE Conference on Computer Vision and Pat- tor and its defect detection based on YOLOX[J]. Sensors,
tern Recognition (CVPR), Honolulu, HI, USA, 21–26 July 2022, 22(16): 6186.
2017, 2017. [28] 无损检测超声试块通用规范: JB/T 8428–2015[S].