Pyramid-based anti-fisheye feature enhancement preprocessing algorithm in torpedo can electrical devices: application in steel rolling process

doi:10.1007/s40436-025-00569-6

Abstract

Abstract: The steel manufacturing industry currently urgently needs highly accurate detection algorithms for electrical connection devices to slow down the time and danger of electrical connections to torpedo cans during high-temperature operations. The fisheye effect and fuzzy features of industrial cameras seriously affect accuracy and effectiveness, hindering the widespread application of object detection algorithms in the manufacturing industry. We propose a feature enhancement preprocessing algorithm for torpedo can electrical devices based on the pyramid structure that resists fisheye effects and serves to detect and locate electrical connection devices. With the aid of this preprocessing algorithm, the detection efficiency and accuracy of state-of-the-art (SOTA) object detection models are significantly improved. Experimental validation confirms the superiority of our method over other SOTA methods. With the application of our preprocessing algorithm, the production capacity of the steel plant increased by 31.8%, and material wastage caused by transportation decreased by 10.9%.

The full text can be downloaded at https://doi.org/10.1007/s40436-025-00569-6

Key words: Anti-fisheye effect (AFE), Feature enhancement, Machine vision, Steel smelting, Torpedo can

Tian-Jie Fu, Shi-Min Liu, Pei-Yu Li, Ruo-Xin Wang. Pyramid-based anti-fisheye feature enhancement preprocessing algorithm in torpedo can electrical devices: application in steel rolling process[J]. Advances in Manufacturing, 2026, 14(2): 329-342.

TrendMD

References

[1] Conrad LF, Oliver ML, Jack RJ et al (2021) Quantification of 6-degree-of-freedom chassis whole-body vibration in mobile heavy vehicles used in the steel making industry. J Low Freq Noise Vib Active Control 31(2):85-104. https://doi.org/10.1260/0263-0923.31.2.85
[2] Fu T, Li P, Liu S (2024) An imbalanced small sample slab defect recognition method based on image generation. J Manuf Process 118:376-388. https://doi.org/10.1016/j.jmapro.2024.03.028
[3] Fu T, Liu S, Li P (2024) Digital twin-driven smelting process management method for converter steelmaking. J Intell Manuf 36:2749-2765. https://doi.org/10.1007/s10845-024-02366-7
[4] Fu T, Li P, Liu S (2025) A method for quality inspection of continuous casting billet based on infrared flaw detection and 2-D image. IEEE Trans Instrum Meas 74:1-14. https://doi.org/10.1109/TIM.2024.3502874
[5] Gao X, Zhang R, You Z et al (2022) Use of hydrogen-rich gas in blast furnace ironmaking of V-bearing titanomagnetite: mass and energy balance calculations. Materials 15(17):6078. https://doi.org/10.3390/ma15176078
[6] Semenov YS, Shumel’chik EL, Gorupakha VV et al (2017) Monitoring blast furnace lining condition during five years of operation. Metallurgist 61(34):291-297. https://doi.org/10.1007/s11015-017-0491-z
[7] Ni J, Shen K, Chen Y et al (2023) An improved SSD-like deep network-based object detection method for indoor scenes. IEEE Trans Instrum Meas 72:1-15. https://doi.org/10.1109/TIM.2023.3244819
[8] Guo J, Chen H, Liu B et al (2023) A system and method for person identification and positioning incorporating object edge detection and scale-invariant feature transformation. Measurement 223:113759. https://doi.org/10.1016/j.measurement.2023.113759
[9] Zhai D, Zhang X, Li X et al (2023) Object detection methods on compressed domain videos: an overview, comparative analysis, and new directions. Measurement 207:112371. https://doi.org/10.1016/j.measurement.2022.112371
[10] Sun Y, Song K, Zhou T et al (2023) A shared method of metal object detection and living object detection based on the quality factor of detection coils for electric vehicle wireless charging. IEEE Trans Instrum Meas 72:1-17. https://doi.org/10.1109/TIM.2023.3277132
[11] Zou Y, Liu C (2023) A light-weight object detection method based on knowledge distillation and model pruning for seam tracking system. Measurement 220:113438. https://doi.org/10.1016/j.measurement.2023.113438
[12] Xie Q, Li D, Yu Z et al (2020) Detecting trees in street images via deep learning with attention module. IEEE Trans Instrum Meas 69(8):5395-5406. https://doi.org/10.1109/TIM.2019.2958580
[13] Dong Z, Liu Y, Feng Y et al (2022) Object detection method for high resolution remote sensing imagery based on convolutional neural networks with optimal object anchor scales. Int J Remote Sens 43(7):2698-2719. https://doi.org/10.1080/01431161.2022.2066487
[14] Ren J, Hu X, Zhu L et al (2023) Deep texture-aware features for camouflaged object detection. IEEE Trans Circuits Syst Video Technol 33(3):1157-1167. https://doi.org/10.1109/TCSVT.2021.3126591
[15] Liu Z, Zhang Z, Tan Y et al (2022) Boosting camouflaged object detection with dual-task interactive transformer. Proceed - Int Confer Pattern Recog 2022:140-146. https://doi.org/10.1109/ICPR56361.2022.9956724
[16] Xue Z, Xue N, Xia GS et al (2019) Learning to calibrate straight lines for fisheye image rectification. Proc IEEE/CVF Conf Comput Vis Pattern Recognit (CVPR) 2019:1643-1651
[17] Fan J, Zhang J, Tao D (2020) SIR: self-supervised image rectification via seeing the same scene from multiple different lenses. IEEE Trans Image Process 32:865-877. https://doi.org/10.1109/TIP.2022.3231087
[18] Chao C, Hsu P, Lee H et al (2020) Self-supervised deep learning for fisheye image rectification. In: Proc 2020 IEEE international conference on acoustics, speech and signal processing (ICASSP) 2020: 2248-2252 https://doi.org/10.1109/icassp40776.2020.9054191
[19] Li T, Tong G, Tang H et al (2020) FisheyeDet: a self-study and contour-based object detector in fisheye images. IEEE Access 8:71739-71751. https://doi.org/10.1109/ACCESS.2020.2987868
[20] Yang CY, Chen HH (2021) Efficient face detection in the fisheye image domain. IEEE Trans Image Process 30:5641-5651. https://doi.org/10.1109/TIP.2021.3087400
[21] Wang W, Wang H, Chen L et al (2023) A novel soft sensor method based on stacked fusion autoencoder with feature enhancement for industrial application. Measure J Int Measure Confeder 221:113491. https://doi.org/10.1016/j.measurement.2023.113491
[22] Li B, Wang T, Zhai Y et al (2022) RFIENet: RGB-thermal feature interactive enhancement network for semantic segmentation of insulator in backlight scenes. Measure J Int Measure Confeder 205:112177. https://doi.org/10.1016/j.measurement.2022.112177
[23] Nirmala DM, Sankar V (2023) Graph based heterogeneous feature extraction for enhanced hardware trojan detection at gate-level using optimized XGBoost algorithm. Measure J Int Measure Confeder 220:113320. https://doi.org/10.1016/j.measurement.2023.113320
[24] Wang J, Yang J, Lu G et al (2023) Adaptively fused attention module for the fabric defect detection. Adv Intell Syst 5(2):2200151. https://doi.org/10.1002/aisy.202200151
[25] Qi Q, Li K, Zheng H et al (2022) SGUIE-Net: semantic attention guided underwater image enhancement with multi-scale perception. IEEE Trans Image Process 31:6816-6830. https://doi.org/10.1109/TIP.2022.3216208
[26] Guo T, Wei Y, Shao H et al (2021) Research on underwater target detection method based on improved MSRCP and YOLOv3. In: Proc 2021 IEEE international conference on mechatronics and automation, ICMA 2021: 1158-1163. https://doi.org/10.1109/ICMA52036.2021.9512827
[27] Hu X, Liu H, Hui Y et al (2022) Transformer feature enhancement network with template update for object tracking. Sensors 22(14):5219. https://doi.org/10.3390/s22145219
[28] Wang CY, Liao HYM, Wu YH et al (2020) CSPNet: a new backbone that can enhance learning capability of CNN. Proc IEEE Conf Comput Vis Pattern Recognit Workshop (CVPR Workshop) 2020:390-391
[29] Fu H, Gong M, Wang C et al (2019) Geometry-consistent generative adversarial networks for one-sided unsupervised domain mapping. In: IEEE conference on computer vision and pattern recognition (CVPR) 2422-2431. https://doi.org/10.1109/CVPR.2019.00253.
[30] Zang X, Li G, Gao W (2022) Multidirection and multiscale pyramid in transformer for video-based pedestrian retrieval. IEEE Trans Industr Inf 18(12):8776-8785. https://doi.org/10.1109/TII.2022.3151766
[31] Koyun OC, Keser RK, Akkaya İB et al (2022) Focus-and-detect: a small object detection framework for aerial images. Signal Process Image Commun 104:116675. https://doi.org/10.1016/j.image.2022.116675
[32] Qi S, Du J, Wu M et al (2022) Underwater small target detection based on deformable convolutional pyramid. In: IEEE international conference on acoustics, speech and signal processing (ICASSP), Singapore, 2022, pp 2784-2788 https://doi.org/10.1109/ICASSP43922.2022.9746575
[33] Sahin O, Ozer S (2021) YOLODrone: improved YOLO architecture for object detection in drone images. In: 2021 44th international conference on telecommunications and signal processing (TSP), pp 361-365. https://doi.org/10.1109/TSP52935.2021.9522653
[34] Liao X, Lv S, Li D et al (2021) Yolov4-mn3 for PCB surface defect detection. Appl Sci 11(24):11701. https://doi.org/10.3390/app112411701
[35] Ren S, He K, Girshick R et al (2015) Faster R-CNN: towards real-time object detection with region proposal networks. arXiv preprint arXiv:1506.01497
[36] Wang Z, Bovik AC, Sheikh HR et al (2004) Image quality assessment: from error visibility to structural similarity. IEEE Trans Image Process 13(4):600-612. https://doi.org/10.1109/TIP.2003.819861
[37] Bashiri FS, LaRose ER, Peissig PL et al (2018) Mcindoor20000: a fully-labeled image dataset to advance indoor objects detection. Data Brief 17:71-75. https://doi.org/10.1016/j.dib.2017.12.047
[38] Jin H, Li L, Su H et al (2024) Learn to enhance the low-light image via a multi-exposure generation and fusion method. J Vis Commun Image Represent 100:104127. https://doi.org/10.1016/j.jvcir.2024.104127
[39] Gao F, Li L, Wang J et al (2023) A lightweight feature distillation and enhancement network for super-resolution remote sensing images. Sensors 23(8):3906. https://doi.org/10.3390/s23083906