Skip to content
2000
Volume 18, Issue 2
  • ISSN: 2352-0965
  • E-ISSN: 2352-0973

Abstract

Aims

A Robust Approach to Object Detection in Images using Improved T_CenterNet Method.

Background

Currently, two-stage image object detectors are the most common method of detecting objects. However, due to the slow processing of two-stage detectors, it is not always possible to apply them in real-time.

Objective

To develop a more robust and effective method for detecting objects in images by building upon and improving the T_CenterNet algorithm.

Methods

In this research paper, we propose CenterNet, a method based on a one-stage detector that takes into consideration the heat map branch of CenterNet, which is the most accurate anchor-free detector. To improve the outcomes of the upcoming image as well as detection efficiency, we disseminate the previously accurate long-term detection using a heatmap and a cascade guiding framework.

Results

On the COCO dataset, our technique obtains detection outcome AP -65.5%, (Small Objects)-33.8, (Medium Objects)-47.0 at 38 frames per second. The results of the study show the superiority of the suggested approach over the conventional one.

Conclusion

In this research, we have introduced an enhanced approach to image object detection using the improved T_CenterNet method. While the original CenterNet algorithm demonstrated superior precision compared to other single-stage algorithms, it suffered from slow detection speed. Our proposed approach, the Hourglass-208 model, built upon CenterNet, significantly improved detection performance. The proposed T_CenterNet algorithm improves upon the original CenterNet by incorporating the Hourglass backbone and feature map fusion techniques, enabling enhanced object detections. Additionally, we applied the Smooth 1 loss function to object size estimation within T_CenterNet, greatly enhancing overall detection performance by improving object size regression.

Loading

Article metrics loading...

/content/journals/raeeng/10.2174/0123520965265205231117073427
2024-01-22
2025-07-10
Loading full text...

Full text loading...

References

  1. GirshickR.B. DonahueJ. DarrellT. MalikJ. Rich feature hierarchies for accurate object detection and semantic segmentation.2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)5805872013
    [Google Scholar]
  2. LiangX. ShenS. XuT. FengJ. YanS. Scale-aware fast R-CNN for pedestrian detection.IEEE Trans. Multimed.201520985996
    [Google Scholar]
  3. LawH. DengJ. CornerNet: Detecting objects as paired keypoints.Int. J. Comput. Vis.2020128364265610.1007/s11263‑019‑01204‑1
    [Google Scholar]
  4. LinT.Y. Microsoft COCO: Common objects in context.European Conference on Computer Vision201474075510.1007/978‑3‑319‑10602‑1_48
    [Google Scholar]
  5. SirishaM. SudhaS.V. Object detection using deep learning centernet model with multi-head external attention mechanism.Int. J. Image Graph.202200245002110.1142/S0219467824500219
    [Google Scholar]
  6. ZhouX. WangD. KrähenbühlP. Objects as points.arXiv:1904.078502019
    [Google Scholar]
  7. LinT.Y. GoyalP. GirshickR. HeK. DollárP. Focal loss for dense object detection.IEEE Trans. Pattern Anal. Mach. Intell.202042231832710.1109/TPAMI.2018.285882630040631
    [Google Scholar]
  8. LiuW. SSD: Single shot multibox detector.Computer Vision – ECCV 20162016213710.1007/978‑3‑319‑46448‑0_2
    [Google Scholar]
  9. LinT-Y. DollárP. GirshickR.B. HeK. HariharanB. BelongieS.J. Feature pyramid networks for object detection.arXiv:1612.031442016936944
    [Google Scholar]
  10. HeK. GkioxariG. DollárP. GirshickR. Mask R-CNN.IEEE Trans. Pattern Anal. Mach. Intell.202042238639710.1109/TPAMI.2018.284417529994331
    [Google Scholar]
  11. RenS. HeK. GirshickR. SunJ. Faster R-CNN: Towards real-time object detection with region proposal networks.IEEE Trans. Pattern Anal. Mach. Intell.20173961137114910.1109/TPAMI.2016.257703127295650
    [Google Scholar]
  12. DaiJ. LiY. HeK. SunJ. R-FCN: Object detection via region-based fully convolutional networks.arXiv:1605.064092016
    [Google Scholar]
  13. YuF. KoltunV. Multi-scale context aggregation by dilated convolutions.Multi-Scale Context Aggregation by Dilated Convolutions2015379387
    [Google Scholar]
  14. HowardA.G. MobileNets: Efficient convolutional neural networks for mobile vision applications.arXiv:1704.048612017
    [Google Scholar]
  15. CaoD. ChenZ. GaoL. An improved object detection algorithm based on multi-scaled and deformable convolutional neural networks.HCIS20201011410.1186/s13673‑020‑00219‑9
    [Google Scholar]
  16. LiY. ChenY. WangN. ZhangZ-X. Scale-aware trident networks for object detection.2019 IEEE/CVF International Conference on Computer Vision (ICCV)20196053606210.1109/ICCV.2019.00615
    [Google Scholar]
  17. HuangL. YangY. DengY. YuY. DenseBox: Unifying landmark localization with end to end object detection.arXiv:1509.048742015
    [Google Scholar]
  18. TianZ. ShenC. ChenH. HeT. FCOS: Fully convolutional one-stage object detection.2019 IEEE/CVF International Conference on Computer Vision (ICCV)20199626963510.1109/ICCV.2019.00972
    [Google Scholar]
  19. ZhouX. ZhuoJ. KrahenbuhlP. Bottom-Up object detection by grouping extreme and center points.2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)201985085910.1109/CVPR.2019.00094
    [Google Scholar]
  20. YuJ. JiangY. WangZ. CaoZ. HuangT. UnitBox.Proceedings of the 24th ACM international conference on Multimedia201651652010.1145/2964284.2967274
    [Google Scholar]
  21. ZhongZ. SunL. HuoQ. An anchor-free region proposal network for Faster R-CNN-based text detection approaches.Int. J. Doc. Anal. Recognit.201922331532710.1007/s10032‑019‑00335‑y
    [Google Scholar]
  22. LiuW. LiaoS. RenW. HuW. YuY. High-level semantic feature detection: A new perspective for pedestrian detection.2019 IEEE/CVF Conf. Comput. Vis. Pattern Recognit.51825191201910.1109/CVPR.2019.00533
    [Google Scholar]
  23. GuoH. YangX. WangN. GaoX. A CenterNet++ model for ship detection in SAR images.Pattern Recognit.202111210778710.1016/j.patcog.2020.107787
    [Google Scholar]
  24. KongT. SunF. LiuH. JiangY. ShiJ. FoveaBox: Beyond Anchor-based Object Detector.ArXiv2019
    [Google Scholar]
  25. WangJ. ChenK. YangS. LoyC.C. LinD. Region proposal by guided anchoring.2019 IEEE/CVF Conf. Comput. Vis. Pattern Recognit.29602969201910.1109/CVPR.2019.00308
    [Google Scholar]
  26. YangZ. LiuS. HuH. WangL. LinS. RepPoints: Point set representation for object detection.2019 IEEE/CVF International Conference on Computer Vision (ICCV)20199656966510.1109/ICCV.2019.00975
    [Google Scholar]
  27. TangC. WuZ. WangS. DengC. LuoL. Industrial object detection method based on improved CenterNet.2021 International Conference on Computer Engineering and Artificial Intelligence (ICCEAI)202112112510.1109/ICCEAI52939.2021.00023
    [Google Scholar]
  28. HeK. ZhangX. RenS. SunJ. Deep residual learning for image recognition.2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)201677077810.1109/CVPR.2016.90
    [Google Scholar]
  29. YuF. WangD. ShelhamerE. DarrellT. Deep layer aggregation.2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition20182403241210.1109/CVPR.2018.00255
    [Google Scholar]
  30. NewellA. YangK. DengJ. Stacked hourglass networks for human pose estimation.ECCV 2016: Computer Vision – ECCV201648349910.1007/978‑3‑319‑46484‑8_29
    [Google Scholar]
  31. ShiX. SuY. TiY. SongT. DaiZ. Improved model based on centernet object detection algorithms.Comput. Eng.202147924025110.19678/j.issn.1000‑3428.0058800
    [Google Scholar]
  32. DaiJ. Deformable convolutional networks.2017 IEEE International Conference on Computer Vision (ICCV)201776477310.1109/ICCV.2017.89
    [Google Scholar]
  33. GirshickR.B. Fast R-CNNComputer Vision and Pattern Recognition (cs.CV)144014482015
    [Google Scholar]
  34. RedmonJ. FarhadiA. YOLOv3: An incremental improvement.arXiv:1804.027672018
    [Google Scholar]
  35. ZhangS. WenL. LeiZ. LiS.Z. RefineDet++: Single-shot refinement neural network for object detection.IEEE Trans. Circ. Syst. Video Tech.202131267468710.1109/TCSVT.2020.2986402
    [Google Scholar]
  36. DuanK. BaiS. XieL. QiH. HuangQ. TianQ. CenterNet: Keypoint triplets for object detection.2019 IEEE/CVF Int. Conf. Comput. Vis.65686577201910.1109/ICCV.2019.00667
    [Google Scholar]
  37. ChenY. ZhangZ. CaoY. WangL. LinS. HuH. RepPoints V2: Verification meets regression for object detection.arXiv:2007.085082020
    [Google Scholar]
  38. LuX. LiB. YueY. LiQ. YanJ. Grid R-CNN.2019 IEEE/CVF Conf. Comput. Vis. Pattern Recognit.735573642018
    [Google Scholar]
  39. DuanK. BaiS. XieL. QiH. HuangQ. TianQ. CenterNet++ for Object Detection.ArXiv2022
    [Google Scholar]
  40. ShenZ. LiuZ. LiJ. JiangY-G. ChenY. XueX. DSOD: Learning deeply supervised object detectors from scratch.2017 IEEE International Conference on Computer Vision (ICCV)20171937194510.1109/ICCV.2017.212
    [Google Scholar]
  41. MaC. LinW. SunL. HuoQ. Robust table detection and structure recognition from heterogeneous document images.Pattern Recognit.202313310900610.1016/j.patcog.2022.109006
    [Google Scholar]
  42. DörrL. BrandtF. NaumannA. PoulsM. TetraPackNet: Four-corner-based object detection in logistics use-cases.Pattern Recognition202154555810.1007/978‑3‑030‑92659‑5_35
    [Google Scholar]
  43. AlbahliS. NawazM. DCNet: DenseNet-77-based CornerNet model for the tomato plant leaf disease detection and classification.Front. Plant Sci.20221395796110.3389/fpls.2022.95796136160977
    [Google Scholar]
  44. QuZ. ZhangR. BaoK. A keypoint‐based object detection method with wide dual‐path backbone network and attention modules.IET Image Process.20211581800181310.1049/ipr2.12152
    [Google Scholar]
  45. JaiswalT. PandeyM. TripathiP. Real time multiple-object detection based on enhanced SSD.2022 Second International Conference on Power, Control and Computing Technologies (ICPC2T)20221510.1109/ICPC2T53885.2022.9776899
    [Google Scholar]
/content/journals/raeeng/10.2174/0123520965265205231117073427
Loading
/content/journals/raeeng/10.2174/0123520965265205231117073427
Loading

Data & Media loading...


  • Article Type:
    Research Article
Keyword(s): anchor free; backbone; deep-learning; Object detection; one stage detection; T_centernet
This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error
Please enter a valid_number test