View in EDS

Heatmap-guided LW-DERT-CS for lightweight and accurate camera calibration.

Saved in:
Bibliographic Details
Title: Heatmap-guided LW-DERT-CS for lightweight and accurate camera calibration.
Authors: Song, Junfang1 (AUTHOR) 284786635@qq.com, Yan, Zhuyang1 (AUTHOR) sun0868sun@163.com, Wang, Shuyu1 (AUTHOR) sywang@xzmu.edu.cn, Ding, Yunyao1 (AUTHOR) dingyunyao163@163.com, Liu, Mingjun1 (AUTHOR) bestfrankliu@outlook.com, Wang, Ziheng1 (AUTHOR) 2976839334@qq.com
Source: Imaging Science Journal. Jul2026, Vol. 74 Issue 5, p552-569. 18p.
Subjects: Camera calibration, Parameter estimation, Artificial neural networks, Optical distortion
Abstract: To address limitations in distortion correction, corner detection accuracy, and parameter estimation stability, this study proposes a heatmap-guided LW-DERT-tiny lightweight camera calibration framework. A convolutional neural network is trained to model and compensate radial distortion. The LW-DERT network is enhanced with a C2f_Faster module and an SE attention mechanism to improve corner candidate detection efficiency and accuracy. Subpixel corner localization and outlier rejection are achieved through a heatmap-guided Gaussian surface fitting strategy, with further refinement based on collinearity constraints. Camera parameters are subsequently estimated using a simplified RANSAC algorithm to enhance stability. Experiments conducted on the GoPro, uEye, and self-built datasets yield minimum reprojection errors of 0.20, 0.19, and 0.22 px, respectively. Compared with recent state-of-the-art methods, the proposed approach reduces the average reprojection error by 11.7% on the self-built dataset, indicating improved robustness and calibration accuracy. [ABSTRACT FROM AUTHOR]
Copyright of Imaging Science Journal is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
FullText Text:
  Availability: 0
Header DbId: egs
DbLabel: Engineering Source
An: 194898345
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Heatmap-guided LW-DERT-CS for lightweight and accurate camera calibration.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Song%2C+Junfang%22">Song, Junfang</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> 284786635@qq.com</i><br /><searchLink fieldCode="AR" term="%22Yan%2C+Zhuyang%22">Yan, Zhuyang</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> sun0868sun@163.com</i><br /><searchLink fieldCode="AR" term="%22Wang%2C+Shuyu%22">Wang, Shuyu</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> sywang@xzmu.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Ding%2C+Yunyao%22">Ding, Yunyao</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> dingyunyao163@163.com</i><br /><searchLink fieldCode="AR" term="%22Liu%2C+Mingjun%22">Liu, Mingjun</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> bestfrankliu@outlook.com</i><br /><searchLink fieldCode="AR" term="%22Wang%2C+Ziheng%22">Wang, Ziheng</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> 2976839334@qq.com</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Imaging+Science+Journal%22">Imaging Science Journal</searchLink>. Jul2026, Vol. 74 Issue 5, p552-569. 18p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Camera+calibration%22">Camera calibration</searchLink><br /><searchLink fieldCode="DE" term="%22Parameter+estimation%22">Parameter estimation</searchLink><br /><searchLink fieldCode="DE" term="%22Artificial+neural+networks%22">Artificial neural networks</searchLink><br /><searchLink fieldCode="DE" term="%22Optical+distortion%22">Optical distortion</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: To address limitations in distortion correction, corner detection accuracy, and parameter estimation stability, this study proposes a heatmap-guided LW-DERT-tiny lightweight camera calibration framework. A convolutional neural network is trained to model and compensate radial distortion. The LW-DERT network is enhanced with a C2f_Faster module and an SE attention mechanism to improve corner candidate detection efficiency and accuracy. Subpixel corner localization and outlier rejection are achieved through a heatmap-guided Gaussian surface fitting strategy, with further refinement based on collinearity constraints. Camera parameters are subsequently estimated using a simplified RANSAC algorithm to enhance stability. Experiments conducted on the GoPro, uEye, and self-built datasets yield minimum reprojection errors of 0.20, 0.19, and 0.22 px, respectively. Compared with recent state-of-the-art methods, the proposed approach reduces the average reprojection error by 11.7% on the self-built dataset, indicating improved robustness and calibration accuracy. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Imaging Science Journal is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=194898345
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1080/13682199.2026.2630579
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 18
        StartPage: 552
    Subjects:
      – SubjectFull: Camera calibration
        Type: general
      – SubjectFull: Parameter estimation
        Type: general
      – SubjectFull: Artificial neural networks
        Type: general
      – SubjectFull: Optical distortion
        Type: general
    Titles:
      – TitleFull: Heatmap-guided LW-DERT-CS for lightweight and accurate camera calibration.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Song, Junfang
      – PersonEntity:
          Name:
            NameFull: Yan, Zhuyang
      – PersonEntity:
          Name:
            NameFull: Wang, Shuyu
      – PersonEntity:
          Name:
            NameFull: Ding, Yunyao
      – PersonEntity:
          Name:
            NameFull: Liu, Mingjun
      – PersonEntity:
          Name:
            NameFull: Wang, Ziheng
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 07
              Text: Jul2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 13682199
          Numbering:
            – Type: volume
              Value: 74
            – Type: issue
              Value: 5
          Titles:
            – TitleFull: Imaging Science Journal
              Type: main
ResultId 1