Application of deep learning for real-time detection, localization, and counting of the malignant invasive weed Solanum rostratum Dunal
Solanum rostratum Dunal (SrD) is a globally harmful invasive weed that has spread widely across many countries, posing a serious threat to agriculture and ecosystem security. A deep learning network model, TrackSolanum, was designed for real-time detection, location, and counting of SrD in the field...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-01-01
|
| Series: | Frontiers in Plant Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fpls.2024.1486929/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832582884329783296 |
|---|---|
| author | Shifeng Du Yashuai Yang Hongbo Yuan Man Cheng |
| author_facet | Shifeng Du Yashuai Yang Hongbo Yuan Man Cheng |
| author_sort | Shifeng Du |
| collection | DOAJ |
| description | Solanum rostratum Dunal (SrD) is a globally harmful invasive weed that has spread widely across many countries, posing a serious threat to agriculture and ecosystem security. A deep learning network model, TrackSolanum, was designed for real-time detection, location, and counting of SrD in the field. The TrackSolanmu network model comprises four modules: detection, tracking, localization, and counting. The detection module uses YOLO_EAND for SrD identification, the tracking module applies DeepSort for multi-target tracking of SrD in consecutive video frames, the localization module determines the position of the SrD through center-of-mass localization, and the counting module counts the plants using a target ID over-the-line invalidation method. The field test results show that for UAV video at a height of 2m, TrackSolanum achieved precision and recall of 0.950 and 0.970, with MOTA and IDF1 scores of 0.826 and 0.960, a counting error rate of 2.438%, and FPS of 17. For UAV video at a height of 3m, the model reached precision and recall of 0.846 and 0.934, MOTA and IDF1 scores of 0.708 and 0.888, a counting error rate of 4.634%, and FPS of 79. Thus, the TrackSolanum supports real-time SrD detection, offering crucial technical support for hazard assessment and precise management of SrD. |
| format | Article |
| id | doaj-art-ceae870ab32a4ff9a7dca5a00a865ca3 |
| institution | Kabale University |
| issn | 1664-462X |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Plant Science |
| spelling | doaj-art-ceae870ab32a4ff9a7dca5a00a865ca32025-01-29T06:46:08ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2025-01-011510.3389/fpls.2024.14869291486929Application of deep learning for real-time detection, localization, and counting of the malignant invasive weed Solanum rostratum DunalShifeng DuYashuai YangHongbo YuanMan ChengSolanum rostratum Dunal (SrD) is a globally harmful invasive weed that has spread widely across many countries, posing a serious threat to agriculture and ecosystem security. A deep learning network model, TrackSolanum, was designed for real-time detection, location, and counting of SrD in the field. The TrackSolanmu network model comprises four modules: detection, tracking, localization, and counting. The detection module uses YOLO_EAND for SrD identification, the tracking module applies DeepSort for multi-target tracking of SrD in consecutive video frames, the localization module determines the position of the SrD through center-of-mass localization, and the counting module counts the plants using a target ID over-the-line invalidation method. The field test results show that for UAV video at a height of 2m, TrackSolanum achieved precision and recall of 0.950 and 0.970, with MOTA and IDF1 scores of 0.826 and 0.960, a counting error rate of 2.438%, and FPS of 17. For UAV video at a height of 3m, the model reached precision and recall of 0.846 and 0.934, MOTA and IDF1 scores of 0.708 and 0.888, a counting error rate of 4.634%, and FPS of 79. Thus, the TrackSolanum supports real-time SrD detection, offering crucial technical support for hazard assessment and precise management of SrD.https://www.frontiersin.org/articles/10.3389/fpls.2024.1486929/fullinvasive plantsSolanum rostratum Dunaldeep learningreal-time detectionlocalizationcounting |
| spellingShingle | Shifeng Du Yashuai Yang Hongbo Yuan Man Cheng Application of deep learning for real-time detection, localization, and counting of the malignant invasive weed Solanum rostratum Dunal Frontiers in Plant Science invasive plants Solanum rostratum Dunal deep learning real-time detection localization counting |
| title | Application of deep learning for real-time detection, localization, and counting of the malignant invasive weed Solanum rostratum Dunal |
| title_full | Application of deep learning for real-time detection, localization, and counting of the malignant invasive weed Solanum rostratum Dunal |
| title_fullStr | Application of deep learning for real-time detection, localization, and counting of the malignant invasive weed Solanum rostratum Dunal |
| title_full_unstemmed | Application of deep learning for real-time detection, localization, and counting of the malignant invasive weed Solanum rostratum Dunal |
| title_short | Application of deep learning for real-time detection, localization, and counting of the malignant invasive weed Solanum rostratum Dunal |
| title_sort | application of deep learning for real time detection localization and counting of the malignant invasive weed solanum rostratum dunal |
| topic | invasive plants Solanum rostratum Dunal deep learning real-time detection localization counting |
| url | https://www.frontiersin.org/articles/10.3389/fpls.2024.1486929/full |
| work_keys_str_mv | AT shifengdu applicationofdeeplearningforrealtimedetectionlocalizationandcountingofthemalignantinvasiveweedsolanumrostratumdunal AT yashuaiyang applicationofdeeplearningforrealtimedetectionlocalizationandcountingofthemalignantinvasiveweedsolanumrostratumdunal AT hongboyuan applicationofdeeplearningforrealtimedetectionlocalizationandcountingofthemalignantinvasiveweedsolanumrostratumdunal AT mancheng applicationofdeeplearningforrealtimedetectionlocalizationandcountingofthemalignantinvasiveweedsolanumrostratumdunal |