Transmission Line Segmentation Solutions for UAV Aerial Photography Based on Improved UNet
The accurate and efficient detection of power lines and towers in aerial drone images with complex backgrounds is crucial for the safety of power grid operations and low-altitude drone flights. In this paper, we propose a new method that enhances the deep learning segmentation model UNet algorithm c...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-04-01
|
| Series: | Drones |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2504-446X/7/4/274 |
| _version_ | 1797605724566585344 |
|---|---|
| author | Min He Liang Qin Xinlan Deng Sihan Zhou Haofeng Liu Kaipei Liu |
| author_facet | Min He Liang Qin Xinlan Deng Sihan Zhou Haofeng Liu Kaipei Liu |
| author_sort | Min He |
| collection | DOAJ |
| description | The accurate and efficient detection of power lines and towers in aerial drone images with complex backgrounds is crucial for the safety of power grid operations and low-altitude drone flights. In this paper, we propose a new method that enhances the deep learning segmentation model UNet algorithm called TLSUNet. We enhance the UNet algorithm by using a lightweight backbone structure to extract the features and then reconstructing them with contextual information features. In this network model, to reduce its parameters and computational complexity, we adopt DFC-GhostNet (Dubbed Full Connected) as the backbone feature extraction network, which is composed of the DFC-GhostBottleneck structure and uses asymmetric convolution to capture long-distance targets in transmission lines, thus enhancing the model’s extraction capability. Additionally, we design a hybrid feature extraction module based on convolution and a transformer to refine deep semantic features and improve the model’s ability to locate towers and transmission lines in complex environments. Finally, we adopt the up-sampling operator CARAFE (Content-Aware Re-Assembly of FEature) to improve segmentation accuracy by enhancing target restoration using contextual neighborhood pixel information correlation under feature decoding. Our experiments on public aerial photography datasets demonstrate that the improved model requires only 8.3% of the original model’s computational effort and has only 21.4% of the original model’s parameters, while achieving a reduction in inference speed delay by 0.012 s. The segmentation metrics also showed significant improvements, with the mIOU improving from 79.75% to 86.46% and the mDice improving from 87.83% to 92.40%. These results confirm the effectiveness of our proposed method. |
| first_indexed | 2024-03-11T05:05:08Z |
| format | Article |
| id | doaj.art-8478954b9ae7468fb22f61951aa2faad |
| institution | Directory Open Access Journal |
| issn | 2504-446X |
| language | English |
| last_indexed | 2024-03-11T05:05:08Z |
| publishDate | 2023-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Drones |
| spelling | doaj.art-8478954b9ae7468fb22f61951aa2faad2023-11-17T18:58:25ZengMDPI AGDrones2504-446X2023-04-017427410.3390/drones7040274Transmission Line Segmentation Solutions for UAV Aerial Photography Based on Improved UNetMin He0Liang Qin1Xinlan Deng2Sihan Zhou3Haofeng Liu4Kaipei Liu5School of Electrical and Automation, Wuhan University, Wuhan 430072, ChinaSchool of Electrical and Automation, Wuhan University, Wuhan 430072, ChinaSchool of Electrical and Automation, Wuhan University, Wuhan 430072, ChinaSchool of Electrical and Automation, Wuhan University, Wuhan 430072, ChinaSchool of Electrical and Automation, Wuhan University, Wuhan 430072, ChinaSchool of Electrical and Automation, Wuhan University, Wuhan 430072, ChinaThe accurate and efficient detection of power lines and towers in aerial drone images with complex backgrounds is crucial for the safety of power grid operations and low-altitude drone flights. In this paper, we propose a new method that enhances the deep learning segmentation model UNet algorithm called TLSUNet. We enhance the UNet algorithm by using a lightweight backbone structure to extract the features and then reconstructing them with contextual information features. In this network model, to reduce its parameters and computational complexity, we adopt DFC-GhostNet (Dubbed Full Connected) as the backbone feature extraction network, which is composed of the DFC-GhostBottleneck structure and uses asymmetric convolution to capture long-distance targets in transmission lines, thus enhancing the model’s extraction capability. Additionally, we design a hybrid feature extraction module based on convolution and a transformer to refine deep semantic features and improve the model’s ability to locate towers and transmission lines in complex environments. Finally, we adopt the up-sampling operator CARAFE (Content-Aware Re-Assembly of FEature) to improve segmentation accuracy by enhancing target restoration using contextual neighborhood pixel information correlation under feature decoding. Our experiments on public aerial photography datasets demonstrate that the improved model requires only 8.3% of the original model’s computational effort and has only 21.4% of the original model’s parameters, while achieving a reduction in inference speed delay by 0.012 s. The segmentation metrics also showed significant improvements, with the mIOU improving from 79.75% to 86.46% and the mDice improving from 87.83% to 92.40%. These results confirm the effectiveness of our proposed method.https://www.mdpi.com/2504-446X/7/4/274transmission line segmentationUAVUNetlight-weighting modelACmixCARAFE |
| spellingShingle | Min He Liang Qin Xinlan Deng Sihan Zhou Haofeng Liu Kaipei Liu Transmission Line Segmentation Solutions for UAV Aerial Photography Based on Improved UNet Drones transmission line segmentation UAV UNet light-weighting model ACmix CARAFE |
| title | Transmission Line Segmentation Solutions for UAV Aerial Photography Based on Improved UNet |
| title_full | Transmission Line Segmentation Solutions for UAV Aerial Photography Based on Improved UNet |
| title_fullStr | Transmission Line Segmentation Solutions for UAV Aerial Photography Based on Improved UNet |
| title_full_unstemmed | Transmission Line Segmentation Solutions for UAV Aerial Photography Based on Improved UNet |
| title_short | Transmission Line Segmentation Solutions for UAV Aerial Photography Based on Improved UNet |
| title_sort | transmission line segmentation solutions for uav aerial photography based on improved unet |
| topic | transmission line segmentation UAV UNet light-weighting model ACmix CARAFE |
| url | https://www.mdpi.com/2504-446X/7/4/274 |
| work_keys_str_mv | AT minhe transmissionlinesegmentationsolutionsforuavaerialphotographybasedonimprovedunet AT liangqin transmissionlinesegmentationsolutionsforuavaerialphotographybasedonimprovedunet AT xinlandeng transmissionlinesegmentationsolutionsforuavaerialphotographybasedonimprovedunet AT sihanzhou transmissionlinesegmentationsolutionsforuavaerialphotographybasedonimprovedunet AT haofengliu transmissionlinesegmentationsolutionsforuavaerialphotographybasedonimprovedunet AT kaipeiliu transmissionlinesegmentationsolutionsforuavaerialphotographybasedonimprovedunet |