Abstract
Transmission tower object detection from aerial image is an important task of power transmission fundamental infrastructure monitoring. However, the geometric variations, large-scale variation, complex background features and truss structure of transmission tower in aerial remote sensing images bring huge challenges to object detection. In this work, we propose a feature enhanced convolution network method (DSA-YOLOv3) to improve the performance of transmission tower detection in aerial images by modifying YOLOv3. The four layer feature extraction and fusion in darknet53 deep residual network is designed to enhance the feature expression ability of the network. Secondly, the DSA enhanced feature extraction module (the module consists of deformable convolution, SPP module, and attention module, which is named DSA module) is proposed. We use deformable convolution to improve the network's generalization ability due to the geometric variations of transmission tower in aerial image. Considering the large-scale variation of the transmission towers, we use the SPP module to increase the effective receptive field of the network. Then an attention module is used to eliminate the interference for the mixed characteristics of target and background environment, which is caused by truss structure of the transmission tower. Finally, we use CIoU as the loss function and DIoU as the non-maximum suppression judgment condition to improve the discrimination of highly overlapping targets. The proposed method was used in the transmission tower data sets to experiment. Experimental results demonstrate that the proposed method's AP, recall rate, and precision are 93.52%, 87.55%, and 96.16%, respectively. Compared with the original YOLOv3, the indexes of our method improved by 3.72%, 2.32%, and 5.91%, respectively.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Shengyin, S.: Application of satellite remote sensing image technology in electric power survey. Jiangxi Build. Mater. 01, 61–62 (2020)
Xi, L., Xiangyu, X.: Estimation method of night time images’electric power consumption based on Boston matrix. Geomat. Inform. Sci. Wuhan Univ. 43(12), 1994–2002 (2018)
Rujun, D.: Research on inspection method of hidden danger of tansmission line Tower foundation based on remote sensing satellinte image. North China Electric Power University (Beijing) (2020)
Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. In: Proceedings of 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, San Diego, CA, vol. 1, pp. 886–893. IEEE (2005)
Lowe, D.: Distinctive image features from scale-invariant keypoints. Int. J. Comput. Vision 60(2), 91–110 (2004)
Bay, H., Ess, A., Tuytelaars, T., et al.: Speeded-up robust features (SURF). Comput. Vis. Image Underst. 110(3), 346–359 (2008)
Zhaoxia, Y., Zhengrong, Z., Chao, T., et al.: Hyperspectral image classification based on the combination of spatial-spectral feature and sparse representation. Acta Gewdaetica et Cartographica Sinica 44(7), 75–781 (2015)
Li, S., Hong, T., Shidong, W., et al.: River extraction from the high resolution remote sensing lmage based on spatially correlated pixels template and adboost algorithm. Acta Geodlactica et Cartographica Sinica 42(3), 344–350 (2013)
Girshick, R., Donahue, J., Darell, T.: Ricth feature hierarchies for accurate object detection and semantic segmentation In: Proceedings of 27th IEEE Conference on Computer Vision and Pattern Recognition, pp. 580–587. IEEE Press, Piscataway (2014)
Wenbin, Y., Chenbo, W., Cui, Y., et al.: A method of aitrcraft recoginition in remote sensing images. Bull. Surv. Mapp. 03, 34–37 (2017)
Kaiming, H., Xiangyu, Z., Shaoqing, R., et al.: Spatial pyramid pooling in deep convolutional networks for visual recognition]. IEEE Trans. Pattern Anal. Mach. Intell. 37(9), 1904–1916 (2015)
Jintao, S.: Research on foreign matter monitoring of power grid with faster R-CNN based on sample expansion. Power Grid Technol. 44(01), 44–51 (2020)
Zhi, Y., Wenhao, O., Xiangze, F., et al.: Smart identification of transmission tower based on high-resolution SAR image and deep learning. Electr. Measur. Instrument. 57(04), 71–77 (2020)
Lin, T.Y., Goyal, P., Girshick, R., et al.: Focal loss for Dense Object Detection. IEEE Trans. Pattern Anal. Mach. Intell. (2017)
Yiqing L.: Research on target detection in high resolution SAR image based on deep learning convolutional neural network. Changsha University of Technology (2019)
Xia, G.S., Bai, X., Ding, J., et al.: DOTA: a largescale dataset for object detection in aerial images. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, pp. 3974–3983. IEEE (2018)
Razakarivony, S., Jurie, F.: Vehicle detection in aerial imagery: a small target detection benchmark. J. Vis. Commun. Image Represent. 34, 187–203 (2016)
Everingham, M., Gool, L.V., Williams, C.K.I., et al.: The pascal visual object classes (VOC) challenge. Int. J. Comput. Vision, 3485–3492 (2010)
Redmon, J., Farhadi, A.: YOLOv3: an incremental improvement: arXiv:1804.02767v1[R/OL]. Cornell University, Ithaca, NY, US: (2018)
Zhou, J., Tian, Y., Yuan, C., et al.: Improved uav opium poppy detection using an updated yolov3 model. Sensors 19(22), 4851 (2019)
Jifeng, D., Haozhi, Q., Yuwen, X., et al.: Deformable convolutional networks. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 764–773 (2017)
Woo, S., Park, J., Lee, J.Y., et al.: CBAM: convolutional block attention module. Springer, Cham (2018)
Zhaohui, Z., Ping, W., Wei, L., et al.: Distance-IoU loss: faster and better learning for bounding box regression. arxiv:1911.08287 (2019)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Zhang, Z., Xie, X., Song, C., Dai, D., Bu, L. (2022). Transmission Tower Detection Algorithm Based on Feature-Enhanced Convolutional Network in Remote Sensing Image. In: Yu, S., et al. Pattern Recognition and Computer Vision. PRCV 2022. Lecture Notes in Computer Science, vol 13536. Springer, Cham. https://doi.org/10.1007/978-3-031-18913-5_43
Download citation
DOI: https://doi.org/10.1007/978-3-031-18913-5_43
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-18912-8
Online ISBN: 978-3-031-18913-5
eBook Packages: Computer ScienceComputer Science (R0)