Deeplabv3+ for extracting Enteromorpha prolifera from drone images
DOI:
https://doi.org/10.54097/hset.v56i.9813Keywords:
deeplabv3 , extraction of enteromorpha prolifera, irregular object segmentationAbstract
The coastal waters of Qingdao have been invaded by Enteromorpha proliferata for 15 consecutive years, causing enormous economic losses and becoming one of the hot spots in marine ecology research in China in recent years. At present, the method used for extracting Enteromorpha prolifera commercially is still manual labeling, which consumes labor and time. In addition, although in-depth learning has developed rapidly in the field of image semantics segmentation, the irregularity of Enteromorpha proliferata and the highly affine transformation brought by the UAV shooting perspective will reduce the accuracy and robustness of in-depth learning methods. To solve this problem, this paper uses Deeplabv3+ semantics segmentation model to extract Enteromorpha prolifera from unmanned aerial vehicle images. Void convolution in Deeplabv3+ can improve the recognition ability of Irregular Enteromorpha prolifera, and spatial pyramid architecture can improve the robustness of the model. Experiments show that the recognition results of the algorithm used in this paper coincide well with those of visual interpretation, and can distinguish between Enteromorpha prolifera and floating objects on the sea such as ships. Deeplabv3+ improves the efficiency of extracting Enteromorpha proliferata and is real-time. Combining with unmanned aerial imagery can reduce the cost of harnessing marine ecological problems.
Downloads
References
Zhou M-J, Liu D-Y, Anderson D M, et al. Introduction to the Special Issue on green tides in the Yellow Sea[J]. Estuarine, Coastal and Shelf Science, 2015, 163: 3-8.
Valiela I, Mcclelland J, Hauxwell J, et al. Macroalgal blooms in shallow estuaries: controls and ecophysiological and ecosystem consequences[J]. Limnology and oceanography, 1997, 42(5part2): 1105-1118.
Fletcher R. The occurrence of “green tides”—a review[J]. Marine benthic vegetation: recent changes and the effects of eutrophication, 1996: 7-43.
Xu F, Gao Z, Ning J, et al. Error analysis on Green Tide monitoring using MODIS data in the Yellow Sea based on GF-1 WFV data[C]. Remote Sensing and Modeling of Ecosystems for Sustainability XIII, 2016: 191-196.
Cui B, Zhang H, Jing W, et al. SRSe-net: Super-resolution-based semantic segmentation network for green tide extraction[J]. Remote Sensing, 2022, 14(3): 710.
Wan X, Wan J, Xu M, et al. Enteromorpha coverage information extraction by 1D-CNN and Bi-LSTM networks considering sample balance from GOCI images[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021, 14: 9306-9317.
Lelong C C, Burger P, Jubelin G, et al. Assessment of unmanned aerial vehicles imagery for quantitative monitoring of wheat crop in small plots[J]. Sensors, 2008, 8(5): 3557-3585.
Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation[C]. Proceedings of the IEEE conference on computer vision and pattern recognition, 2015: 3431-3440.
Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation[C]. Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 18, 2015: 234-241.
Badrinarayanan V, Kendall A, Cipolla R. Segnet: A deep convolutional encoder-decoder architecture for image segmentation[J]. IEEE transactions on pattern analysis and machine intelligence, 2017, 39(12): 2481-2495.
Zhao H, Shi J, Qi X, et al. Pyramid scene parsing network[C]. Proceedings of the IEEE conference on computer vision and pattern recognition, 2017: 2881-2890.
Chen L-C, Papandreou G, Kokkinos I, et al. Semantic image segmentation with deep convolutional nets and fully connected crfs[J]. arXiv preprint arXiv:1412.7062, 2014.
Chen L-C, Papandreou G, Kokkinos I, et al. Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs[J]. IEEE transactions on pattern analysis and machine intelligence, 2017, 40(4): 834-848.
Chen L-C, Papandreou G, Schroff F, et al. Rethinking atrous convolution for semantic image segmentation[J]. arXiv preprint arXiv:1706.05587, 2017.
Yu F, Koltun V. Multi-scale context aggregation by dilated convolutions[J]. arXiv preprint arXiv:1511.07122, 2015.
He K, Zhang X, Ren S, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition[J]. IEEE transactions on pattern analysis and machine intelligence, 2015, 37(9): 1904-1916.
Chollet F. Xception: Deep learning with depthwise separable convolutions[C]. Proceedings of the IEEE conference on computer vision and pattern recognition, 2017: 1251-1258.
Chen L-C, Zhu Y, Papandreou G, et al. Encoder-decoder with atrous separable convolution for semantic image segmentation[C]. Proceedings of the European conference on computer vision (ECCV), 2018: 801-818.
Krizhevsky A, Sutskever I, Hinton G. Imagenet classification with deep convolutional networks[C]. Proceedings of the 26th Annual Conference on Neural Information Processing Systems (NIPS): 1106-1114.
Li K, Zhang Y-S, Tong X-C, et al. An edge detection algorithm based on anisotropy and edge strength correction factor[J]. Computer Engineering & Science, 43(07): 1256.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
