Abstract
The use of high-spatial-resolution remote sensing image technology on mobile and embedded equipment is an important and effective way for emergency rescue and evaluation decision-makers to quickly and accurately detect landslide areas. Deep learning-based landslide detection models include one-stage and two-stage models. The two-stage landslide detection models are slower. The one-stage landslide detection models are faster but less accurate. Both types of detection models have many parameters. This research aims to improve the speed, accuracy, and parameters of landslide detection models. A you only look once-small attention (YOLO-SA) landslide detection model is proposed. YOLO-SA is an improved version of the one-stage detection model YOLOv4. First, the group convolution (Gconv) and ghost bottleneck (G-bneck) residual modules are used to replace the convolution components and residual module consisting of standard convolution. The purpose is to reduce the parameters of the model. Then, on this basis, an attention mechanism is added to improve the detection accuracy of the model. Finally, the position of the attention mechanism is adjusted to determine the framework of YOLO-SA. Qiaojia and Ludian counties in Yunnan Province, China, are used as the study area to acquire three-channel (red, green, blue) historical landslide optical remote sensing images from Google Earth, with a total of 1818 images, for training the model. YOLO-SA is compared with 11 advanced models, including Faster-RCNN, 3 types of EfficientDet, 2 types of Centernet, SSD-efficient, and 4 types of YOLOv4 models. The results show that the number of YOLO-SA parameters is reduced to 1.472 mb compared to EfficientDet-D0; the accuracy is improved to 94.08% compared to Centernet-hourglass; and the speed is up to 42 f/s. In addition, the effectiveness of the YOLO-SA model for potential landslide detection is verified, with an F1 score of 90.65%.
Similar content being viewed by others
Availability of data and material
The datasets used or analyzed during the current study are available from the corresponding author on reasonable request.
Code availability
The code used during the current study is available from the corresponding author on reasonable request.
References
Amatya P, Kirschbaum D, Stanley T (2019) Use of very high-resolution optical data for landslide mapping and susceptibility analysis along the Karnali Highway, Nepal. Remote Sens 11:2284. https://doi.org/10.3390/rs11192284
Bochkovskiy A, Wang CY, Liao HYM (2020) YOLOv4: optimal speed and accuracy of object detection. arXiv preprint arXiv:2004.10934
Cao Y, Xu J, Lin S, Wei F, Hu H (2019) GCNet: non-local networks meet squeeze-excitation networks and beyond. arXiv preprint arXiv:1904.11492
Cheng G, Guo L, Zhao T, Han J, Li H, Fang J (2013) Automatic landslide detection from remote-sensing imagery using a scene classification method based on BoVW and pLSA. Int J Remote Sens 34:45–59. https://doi.org/10.1080/01431161.2012.705443
Cheng G, Li R, Lang C, Han J (2021) Task-wise attention guided part complementary learning for few-shot image classification. Sci China Inform Sci 64:120104
Di Napoli M et al (2020) Machine learning ensemble modelling as a tool to improve landslide susceptibility mapping reliability. Landslides 17:1897–1914. https://doi.org/10.1007/s10346-020-01392-9
Du S, Zang P, Zang B, Xu H (2021) Weak and occluded vehicle detection in complex infrared environment based on improved YOLOv4. IEEE Access 9:25671–25680. https://doi.org/10.1109/ACCESS.2021.3057723
Fu CY, Liu W, Ranga A, Tyagi A, Berg AC (2017) DSSD: deconvolutional single shot detector. arXiv preprint arXiv:170106659
Galli M, Ardizzone F, Cardinali M, Guzzetti F, Reichenbach PJG (2008) Comparing landslide inventory maps. Geomorphology 94:268–289. https://doi.org/10.1016/j.geomorph.2006.09.023
Gu T, Li J, Wang M, Duan P (2021) Landslide susceptibility assessment in Zhenxiong County of China based on geographically weighted logistic regression model. Geocarto Int:1–23. https://doi.org/10.1080/10106049.2021.1903571
He K, Zhang X, Ren S, Sun J (2014) Spatial pyramid pooling in deep convolutional networks for visual recognition. IEEE Trans Pattern Anal 37:1904–1916
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. Proceedings of the IEEE conference on computer vision and pattern recognition, p 770-778. https://doi.org/10.1109/CVPR.2016.90
He K, Gkioxari G, Dollár P, Girshick R Mask R-CNN (2017) In: arXiv preprint arXiv:1703.06870
Hong et al (2019) Improved faster R-CNN with multiscale feature fusion and homography augmentation for vehicle detection in remote sensing images. IEEE Geosci Remote Sens Lett 16:1761–1765. https://doi.org/10.1109/LGRS.2019.2909541
Howard AG et al. (2017) MobileNets: efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:170404861
Howard A et al. (2019) Searching for MobileNetV3. Proceedings of the IEEE conference on computer vision and pattern recognition, pp 1314-1324
Hu J, Shen L, Albanie S, Sun G, Wu E (2018) Squeeze-and-excitation networks. IEEE Trans Pattern Anal Mach Intell:7132–7141
Ji S, Yu D, Shen C, Li W, Xu Q (2020) Landslide detection from an open satellite imagery and digital elevation model dataset using attention boosted convolutional neural networks. Landslides 17:1337–1352. https://doi.org/10.1007/s10346-020-01353-2
Li T-T, Pei X-J, Huang R-Q, Jin L-D (2016) The formation and evolution of the Qiaojia pull-apart basin, North Xiaojiang Fault Zone, Southwest China. J Mt Sci Engl 13:1096–1106. https://doi.org/10.1007/s11629-015-3778-1
Li X, Wang W, Hu X, Yang J (2019) Selective Kernel Networks. Proceedings of the IEEE conference on computer vision and pattern recognition, p 510-519
Lin TY, Goyal P, Girshick R, He K, Dollár P (2017) Focal loss for dense object detection. IEEE Trans Pattern Anal:2999–3007
Liu W, Anguelov D, Erhan D (2016) SSD: Single shot multibox detector. arXiv preprint arXiv:151202325: 21-37
Loshchilov I, Hutter F (2016) SGDR: stochastic gradient descent with warm restarts. arXiv preprint arXiv:1608.03983
Ma H, Liu Y, Ren Y, Yu J (2020) Detection of collapsed buildings in post-earthquake remote sensing images based on the improved YOLOv3. Remote Sens 12:44. https://doi.org/10.3390/rs12010044
Maxwell AE, Pourmohammadi P, Poyner JD (2020) Mapping the topographic features of mining-related valley fills using mask R-CNN deep learning and digital elevation data. Remote Sens 12:547. https://doi.org/10.3390/rs12010044
Messeri A, Morabito M, Messeri G, Brandani G, Petralli M, Natali F, Grifoni D, Crisci A, Gensini G, Orlandini S (2015) Weather-related flood and landslide damage: a risk index for Italian Regions. PLoS One 10:e0144468
Polyak BT, Juditsky AB (1992) Acceleration of stochastic approximation by averaging. SIAM J Control Optim 30:838–855. https://doi.org/10.1137/0330046
Ramachandran P, Zoph B, Le QV (2017) Swish: a self-gated activation function. arXiv preprint arXiv:1710.05941
Redmon J, Farhadi A (2017) YOLO9000: better, faster, stronger. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 6517-6525
Redmon J, Farhadi A (2018) YOLOv3: an incremental improvement. arXiv preprint arXiv:180402767
Redmon J, Divvala S, Girshick R, Farhadi A (2015) You only look once: unified, real-time object detection. Proceedings of the IEEE conference on computer vision and pattern recognition, p 779-788
Ren S, He K, Girshick R, Sun J (2017) Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Trans Pattern Anal 39:1137–1149
Rezatofighi H, Tsoi N, Gwak JY, Sadeghian A, Savarese S (2019) Generalized intersection over union: a metric and a loss for bounding box regression. In: Proceedings of the IEEE conference on computer vision and pattern recognition
Roy AG, Navab N, Wachinger C (2018) Concurrent spatial and channel squeeze & excitation in fully convolutional networks. International conference on medical image computing and computer-assisted intervention. Springer, Cham, pp 421–429
Srivastava RK, Greff K, Schmidhuber J (2015) Training very deep networks. arXiv preprint arXiv:150500387
Tan M, Pang R, Le QV (2019) EfficientDet: scalable and efficient object detection. arXiv preprint arXiv:191109070
Wang X, Girshick R, Gupta A, He K (2017) Non-local neural networks. Proceedings of the IEEE conference on computer vision and pattern recognition, p 7794-7803
Wang H, Liu S, Xu W, Yan L, Qu X, Xie W-C (2020) Numerical investigation on the sliding process and deposit feature of an earthquake-induced landslide: a case study. Landslides 17:2671–2682. https://doi.org/10.1007/s10346-020-01446-y
Xu Z, Chen Y, Yang F, Chu T, Zhou H (2020) A postearthquake multiple scene recognition model based on classical SSD method and transfer learning. ISPRS Int J Geo-Inf 9:238. https://doi.org/10.3390/ijgi9040238
Yang Y, Deng H (2020) GC-YOLOv3: you only look once with global context block. Electronics 9:1235. https://doi.org/10.3390/electronics9081235
Yang Y, Yang J, Xu C, Xu C, Song C (2019) Local-scale landslide susceptibility mapping using the B-GeoSVC model. Landslides 16:1301–1312. https://doi.org/10.1007/s10346-019-01174-y
Zhang X, Zou Y, Wei S (2017) Dilated convolution neural network with LeakyReLU for environmental sound classification. In: Digit Signal Process, pp 1-5
Zhang H et al. (2020) ResNeSt: split-attention networks. arXiv preprint arXiv:2004.08955
Zheng Z, Wang P, Liu W, Li J, Ren D (2020) Distance-IoU loss: faster and better learning for bounding box regression. In: Distance-IoU loss: faster and better learning for bounding box regression, pp 12993-13000
Zhou X, Wang D, Krhenbühl P (2019) Objects as points. arXiv preprint arXiv:191109070
Acknowledgements
We thank all the authors for their great contribution to this study. We thank Zhoujie Luo and Peng Lai for their help in the experimental data collection and labeling process. Thanks for the valuable landslide data provided by the Yunnan Geological Disaster Department.
Funding
This research was funded by the National Natural Science Foundation of China (No. 41961061) and Yunnan Fundamental Research Projects (grant NO. 202001AT070057).
Author information
Authors and Affiliations
Contributions
Conceptualization: [Jia Li], [Ping Duan]; methodology: [Libo Cheng], [Jia Li]; formal analysis and investigation: [Libo Cheng]; writing - original draft preparation: [Libo Cheng]; writing - review and editing: [Libo Cheng], [Jia Li]; funding acquisition: [Jia Li]; resources: [Mingguo Wang], [Ping Duan]; supervision: [Jia Li], [Ping Duan]; accuracy evaluation: [Mingguo Wang], [Ping Duan].
Corresponding authors
Ethics declarations
Ethics approval
Not applicable.
Consent to participate
Not applicable.
Consent for publication
All authors have read and agreed to the published version of the manuscript.
Competing interests
The authors declare that they have no competing interests.
Rights and permissions
About this article
Cite this article
Cheng, L., Li, J., Duan, P. et al. A small attentional YOLO model for landslide detection from satellite remote sensing images. Landslides 18, 2751–2765 (2021). https://doi.org/10.1007/s10346-021-01694-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10346-021-01694-6