Automatic Estimation Of Rock Quality Designation Based On An Improved YOLOv5.
Saved in:
| Title: | Automatic Estimation Of Rock Quality Designation Based On An Improved YOLOv5. |
|---|---|
| Authors: | Fu, Dong1 (AUTHOR) xd_fu@hhu.edu.cn, Su, Chao1 (AUTHOR), Li, Xiangru1 (AUTHOR) |
| Source: | Rock Mechanics & Rock Engineering. Apr2024, Vol. 57 Issue 4, p3043-3061. 19p. |
| Subjects: | Drill cores, Convolutional neural networks, Core drilling, Geology databases, Image databases |
| Abstract: | Rock quality designation (RQD) characteristics for assessing the degree of rock mass fracture make it a key parameter in rock grading or other rating systems. Traditional core characterization relies on subjective manual visual inspection by geologists. Currently, convolutional neural networks are used in borehole images to classify intact and nonintact cores in core rows for automatic RQD estimation. Classification networks cannot predict the exact locations of the intact cores, and drill core characterization is not intuitive. Alternatively, an attention mechanism combining channel and spatial attention modules is proposed to improve the YOLOv5 algorithm for drill core characterization. The model was trained on 657 artificial core tray images generated by the developed preprocessor to accurately predict the bounding boxes of the intact cores on the row centerline, and the automatic RQD calculation of the row was implemented with the developed postprocessing program. Our method performed RQD estimation on 602 new granite rows and 180 new quartz sandstone rows, with average error rates of 1.27% and 1.12%, respectively. It processed 50 m of cores on average in 1 s on a GPU. Furthermore, this method provides an innovative method for automatically processing and quantifying geological image databases. Highlights: Drill core images are automatically analyzed to estimate the RQD of the rows. An attention mechanism combining channel and spatial attention modules is proposed. YOLOv5 is combined with an attention mechanism for detecting intact cores. The proposed method is tested on granite and quartz sandstone images, yielding an average error rate of 1.24%. [ABSTRACT FROM AUTHOR] |
| Copyright of Rock Mechanics & Rock Engineering is the property of Springer Nature 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 |
|
Full text is not displayed to guests.
Login for full access.
|
|
| FullText | Links: – Type: pdflink Text: Availability: 1 |
|---|---|
| Header | DbId: egs DbLabel: Engineering Source An: 177351140 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Automatic Estimation Of Rock Quality Designation Based On An Improved YOLOv5. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Fu%2C+Dong%22">Fu, Dong</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> xd_fu@hhu.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Su%2C+Chao%22">Su, Chao</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Li%2C+Xiangru%22">Li, Xiangru</searchLink><relatesTo>1</relatesTo> (AUTHOR) – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22Rock+Mechanics+%26+Rock+Engineering%22">Rock Mechanics & Rock Engineering</searchLink>. Apr2024, Vol. 57 Issue 4, p3043-3061. 19p. – Name: Subject Label: Subjects Group: Su Data: <searchLink fieldCode="DE" term="%22Drill+cores%22">Drill cores</searchLink><br /><searchLink fieldCode="DE" term="%22Convolutional+neural+networks%22">Convolutional neural networks</searchLink><br /><searchLink fieldCode="DE" term="%22Core+drilling%22">Core drilling</searchLink><br /><searchLink fieldCode="DE" term="%22Geology+databases%22">Geology databases</searchLink><br /><searchLink fieldCode="DE" term="%22Image+databases%22">Image databases</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Rock quality designation (RQD) characteristics for assessing the degree of rock mass fracture make it a key parameter in rock grading or other rating systems. Traditional core characterization relies on subjective manual visual inspection by geologists. Currently, convolutional neural networks are used in borehole images to classify intact and nonintact cores in core rows for automatic RQD estimation. Classification networks cannot predict the exact locations of the intact cores, and drill core characterization is not intuitive. Alternatively, an attention mechanism combining channel and spatial attention modules is proposed to improve the YOLOv5 algorithm for drill core characterization. The model was trained on 657 artificial core tray images generated by the developed preprocessor to accurately predict the bounding boxes of the intact cores on the row centerline, and the automatic RQD calculation of the row was implemented with the developed postprocessing program. Our method performed RQD estimation on 602 new granite rows and 180 new quartz sandstone rows, with average error rates of 1.27% and 1.12%, respectively. It processed 50 m of cores on average in 1 s on a GPU. Furthermore, this method provides an innovative method for automatically processing and quantifying geological image databases. Highlights: Drill core images are automatically analyzed to estimate the RQD of the rows. An attention mechanism combining channel and spatial attention modules is proposed. YOLOv5 is combined with an attention mechanism for detecting intact cores. The proposed method is tested on granite and quartz sandstone images, yielding an average error rate of 1.24%. [ABSTRACT FROM AUTHOR] – Name: AbstractSuppliedCopyright Label: Group: Ab Data: <i>Copyright of Rock Mechanics & Rock Engineering is the property of Springer Nature 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=177351140 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1007/s00603-023-03729-x Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 19 StartPage: 3043 Subjects: – SubjectFull: Drill cores Type: general – SubjectFull: Convolutional neural networks Type: general – SubjectFull: Core drilling Type: general – SubjectFull: Geology databases Type: general – SubjectFull: Image databases Type: general Titles: – TitleFull: Automatic Estimation Of Rock Quality Designation Based On An Improved YOLOv5. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Fu, Dong – PersonEntity: Name: NameFull: Su, Chao – PersonEntity: Name: NameFull: Li, Xiangru IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 04 Text: Apr2024 Type: published Y: 2024 Identifiers: – Type: issn-print Value: 07232632 Numbering: – Type: volume Value: 57 – Type: issue Value: 4 Titles: – TitleFull: Rock Mechanics & Rock Engineering Type: main |
| ResultId | 1 |
