Adaptive spatial-channel feature fusion and self-calibrated convolution for early maize seedlings counting in UAV images
Accurate counting of crop plants is essential for agricultural science, particularly for yield forecasting, field management, and experimental studies. Traditional methods are labor-intensive and prone to errors. Unmanned Aerial Vehicle (UAV) technology offers a promising alternative; however, varyi...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-02-01
|
| Series: | Frontiers in Plant Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fpls.2024.1496801/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832548281656279040 |
|---|---|
| author | Zhenyuan Sun Zhenyuan Sun Zhenyuan Sun Zhi Yang Zhi Yang Yimin Ding Yimin Ding Boyan Sun Boyan Sun Saiju Li Saiju Li Zhen Guo Zhen Guo Lei Zhu Lei Zhu |
| author_facet | Zhenyuan Sun Zhenyuan Sun Zhenyuan Sun Zhi Yang Zhi Yang Yimin Ding Yimin Ding Boyan Sun Boyan Sun Saiju Li Saiju Li Zhen Guo Zhen Guo Lei Zhu Lei Zhu |
| author_sort | Zhenyuan Sun |
| collection | DOAJ |
| description | Accurate counting of crop plants is essential for agricultural science, particularly for yield forecasting, field management, and experimental studies. Traditional methods are labor-intensive and prone to errors. Unmanned Aerial Vehicle (UAV) technology offers a promising alternative; however, varying UAV altitudes can impact image quality, leading to blurred features and reduced accuracy in early maize seedling counts. To address these challenges, we developed RC-Dino, a deep learning methodology based on DINO, specifically designed to enhance the precision of seedling counts from UAV-acquired images. RC-Dino introduces two innovative components: a novel self-calibrating convolutional layer named RSCconv and an adaptive spatial feature fusion module called ASCFF. The RSCconv layer improves the representation of early maize seedlings compared to non-seedling elements within feature maps by calibrating spatial domain features. The ASCFF module enhances the discriminability of early maize seedlings by adaptively fusing feature maps extracted from different layers of the backbone network. Additionally, transfer learning was employed to integrate pre-trained weights with RSCconv, facilitating faster convergence and improved accuracy. The efficacy of our approach was validated using the Early Maize Seedlings Dataset (EMSD), comprising 1,233 annotated images of early maize seedlings, totaling 83,404 individual annotations. Testing on this dataset demonstrated that RC-Dino outperformed existing models, including DINO, Faster R-CNN, RetinaNet, YOLOX, and Deformable DETR. Specifically, RC-Dino achieved improvements of 16.29% in Average Precision (AP) and 8.19% in Recall compared to the DINO model. Our method also exhibited superior coefficient of determination (R²) values across different datasets for seedling counting. By integrating RSCconv and ASCFF into other detection frameworks such as Faster R-CNN, RetinaNet, and Deformable DETR, we observed enhanced detection and counting accuracy, further validating the effectiveness of our proposed method. These advancements make RC-Dino particularly suitable for accurate early maize seedling counting in the field. The source code for RSCconv and ASCFF is publicly available at https://github.com/collapser-AI/RC-Dino, promoting further research and practical applications. |
| format | Article |
| id | doaj-art-9db0736ff90742268cec300cc86f8213 |
| institution | Kabale University |
| issn | 1664-462X |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Plant Science |
| spelling | doaj-art-9db0736ff90742268cec300cc86f82132025-02-03T06:33:24ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2025-02-011510.3389/fpls.2024.14968011496801Adaptive spatial-channel feature fusion and self-calibrated convolution for early maize seedlings counting in UAV imagesZhenyuan Sun0Zhenyuan Sun1Zhenyuan Sun2Zhi Yang3Zhi Yang4Yimin Ding5Yimin Ding6Boyan Sun7Boyan Sun8Saiju Li9Saiju Li10Zhen Guo11Zhen Guo12Lei Zhu13Lei Zhu14School of Civil and Hydraulic Engineering, Ningxia University, Yinchuan, ChinaKey Laboratory of Digital Water Governance for Yellow River Water Network, Yinchuan, ChinaField Scientific Observation and Research Station of Agricultural Irrigation in Ningxia Diversion Yellow Irrigation District, Ministry of Water Resources, Yinchuan, ChinaField Scientific Observation and Research Station of Agricultural Irrigation in Ningxia Diversion Yellow Irrigation District, Ministry of Water Resources, Yinchuan, ChinaThe Scientific Research Institute of the Water Conservancy of Ningxia, Yinchuan, ChinaSchool of Civil and Hydraulic Engineering, Ningxia University, Yinchuan, ChinaKey Laboratory of Digital Water Governance for Yellow River Water Network, Yinchuan, ChinaSchool of Civil and Hydraulic Engineering, Ningxia University, Yinchuan, ChinaKey Laboratory of Digital Water Governance for Yellow River Water Network, Yinchuan, ChinaSchool of Civil and Hydraulic Engineering, Ningxia University, Yinchuan, ChinaKey Laboratory of Digital Water Governance for Yellow River Water Network, Yinchuan, ChinaSchool of Civil and Hydraulic Engineering, Ningxia University, Yinchuan, ChinaKey Laboratory of Digital Water Governance for Yellow River Water Network, Yinchuan, ChinaSchool of Civil and Hydraulic Engineering, Ningxia University, Yinchuan, ChinaKey Laboratory of Digital Water Governance for Yellow River Water Network, Yinchuan, ChinaAccurate counting of crop plants is essential for agricultural science, particularly for yield forecasting, field management, and experimental studies. Traditional methods are labor-intensive and prone to errors. Unmanned Aerial Vehicle (UAV) technology offers a promising alternative; however, varying UAV altitudes can impact image quality, leading to blurred features and reduced accuracy in early maize seedling counts. To address these challenges, we developed RC-Dino, a deep learning methodology based on DINO, specifically designed to enhance the precision of seedling counts from UAV-acquired images. RC-Dino introduces two innovative components: a novel self-calibrating convolutional layer named RSCconv and an adaptive spatial feature fusion module called ASCFF. The RSCconv layer improves the representation of early maize seedlings compared to non-seedling elements within feature maps by calibrating spatial domain features. The ASCFF module enhances the discriminability of early maize seedlings by adaptively fusing feature maps extracted from different layers of the backbone network. Additionally, transfer learning was employed to integrate pre-trained weights with RSCconv, facilitating faster convergence and improved accuracy. The efficacy of our approach was validated using the Early Maize Seedlings Dataset (EMSD), comprising 1,233 annotated images of early maize seedlings, totaling 83,404 individual annotations. Testing on this dataset demonstrated that RC-Dino outperformed existing models, including DINO, Faster R-CNN, RetinaNet, YOLOX, and Deformable DETR. Specifically, RC-Dino achieved improvements of 16.29% in Average Precision (AP) and 8.19% in Recall compared to the DINO model. Our method also exhibited superior coefficient of determination (R²) values across different datasets for seedling counting. By integrating RSCconv and ASCFF into other detection frameworks such as Faster R-CNN, RetinaNet, and Deformable DETR, we observed enhanced detection and counting accuracy, further validating the effectiveness of our proposed method. These advancements make RC-Dino particularly suitable for accurate early maize seedling counting in the field. The source code for RSCconv and ASCFF is publicly available at https://github.com/collapser-AI/RC-Dino, promoting further research and practical applications.https://www.frontiersin.org/articles/10.3389/fpls.2024.1496801/fullearly maize seedlings countingadaptive spatial-channel feature fusionself-calibrated convolutionRC-Dinodeep learning |
| spellingShingle | Zhenyuan Sun Zhenyuan Sun Zhenyuan Sun Zhi Yang Zhi Yang Yimin Ding Yimin Ding Boyan Sun Boyan Sun Saiju Li Saiju Li Zhen Guo Zhen Guo Lei Zhu Lei Zhu Adaptive spatial-channel feature fusion and self-calibrated convolution for early maize seedlings counting in UAV images Frontiers in Plant Science early maize seedlings counting adaptive spatial-channel feature fusion self-calibrated convolution RC-Dino deep learning |
| title | Adaptive spatial-channel feature fusion and self-calibrated convolution for early maize seedlings counting in UAV images |
| title_full | Adaptive spatial-channel feature fusion and self-calibrated convolution for early maize seedlings counting in UAV images |
| title_fullStr | Adaptive spatial-channel feature fusion and self-calibrated convolution for early maize seedlings counting in UAV images |
| title_full_unstemmed | Adaptive spatial-channel feature fusion and self-calibrated convolution for early maize seedlings counting in UAV images |
| title_short | Adaptive spatial-channel feature fusion and self-calibrated convolution for early maize seedlings counting in UAV images |
| title_sort | adaptive spatial channel feature fusion and self calibrated convolution for early maize seedlings counting in uav images |
| topic | early maize seedlings counting adaptive spatial-channel feature fusion self-calibrated convolution RC-Dino deep learning |
| url | https://www.frontiersin.org/articles/10.3389/fpls.2024.1496801/full |
| work_keys_str_mv | AT zhenyuansun adaptivespatialchannelfeaturefusionandselfcalibratedconvolutionforearlymaizeseedlingscountinginuavimages AT zhenyuansun adaptivespatialchannelfeaturefusionandselfcalibratedconvolutionforearlymaizeseedlingscountinginuavimages AT zhenyuansun adaptivespatialchannelfeaturefusionandselfcalibratedconvolutionforearlymaizeseedlingscountinginuavimages AT zhiyang adaptivespatialchannelfeaturefusionandselfcalibratedconvolutionforearlymaizeseedlingscountinginuavimages AT zhiyang adaptivespatialchannelfeaturefusionandselfcalibratedconvolutionforearlymaizeseedlingscountinginuavimages AT yiminding adaptivespatialchannelfeaturefusionandselfcalibratedconvolutionforearlymaizeseedlingscountinginuavimages AT yiminding adaptivespatialchannelfeaturefusionandselfcalibratedconvolutionforearlymaizeseedlingscountinginuavimages AT boyansun adaptivespatialchannelfeaturefusionandselfcalibratedconvolutionforearlymaizeseedlingscountinginuavimages AT boyansun adaptivespatialchannelfeaturefusionandselfcalibratedconvolutionforearlymaizeseedlingscountinginuavimages AT saijuli adaptivespatialchannelfeaturefusionandselfcalibratedconvolutionforearlymaizeseedlingscountinginuavimages AT saijuli adaptivespatialchannelfeaturefusionandselfcalibratedconvolutionforearlymaizeseedlingscountinginuavimages AT zhenguo adaptivespatialchannelfeaturefusionandselfcalibratedconvolutionforearlymaizeseedlingscountinginuavimages AT zhenguo adaptivespatialchannelfeaturefusionandselfcalibratedconvolutionforearlymaizeseedlingscountinginuavimages AT leizhu adaptivespatialchannelfeaturefusionandselfcalibratedconvolutionforearlymaizeseedlingscountinginuavimages AT leizhu adaptivespatialchannelfeaturefusionandselfcalibratedconvolutionforearlymaizeseedlingscountinginuavimages |