Cloud Detection in High-Resolution Remote Sensing Images Using Multi-features of Ground Objects
- 84 Downloads
The existence of clouds in high-resolution remote sensing images influences target recognition and feature classification. Therefore, finding areas covered with clouds is an important preprocessing step in remote sensing image applications. This paper proposes a cloud detection method for satellite images with high resolution using ground objects’ multi-features, such as color, texture, and shape. First, the highly reflective areas are extracted from the image using the minimum cross entropy threshold method. Second, the multi-scale image decomposition based on domain transform filter extracts the texture features of ground objects. Finally, based on the shape features, regular-shaped artificial ground objects are removed to further improve cloud detection accuracy. The experimental results show that the proposed method not only improves the overall accuracy rate but also reduces the false positive rate compared to the classical traditional cloud detection methods. The method is suitable for cloud detection in high-resolution remote sensing images with complex ground objects.
KeywordsCloud detection Multi-scale decomposition Domain transform filter Regular-shaped artificial ground objects
This work is supported in part by the National Nature Science Foundation of China under Grants 61571345, Yangtse Rive Scholar Bonus Schemes and Ten Thousand Talent Program.
Compliance with Ethical Standards
All authors have participated in (a) conception and design, or analysis and interpretation of the data and (b) drafting the article or revising it critically for important intellectual content.
Conflict of Interest
The authors have no affiliation with any organization with a direct or indirect financial interest in the subject matter discussed in the manuscript.
- An Z, Shi Z (2017) Scene learning for cloud detection on remote-sensing images. IEEE J Sel Top Appl 8(8):4206–4222Google Scholar
- Gastal ES, Oliveira M (2011) Domain transform for edge-aware image and video processing. ACM Trans Graph 30(2011):69Google Scholar
- Hou S, Sun W, Zheng X (2014) Summary of cloud detection methods for remote sensing images. Space Electron Technol 2014(3):68–76Google Scholar
- Jedlovec G, Haines S (2007) Spatial and temporal varying thresholds for cloud detection in satellite imagery geoscience and remote sensing symposium, 2007. IGARSS 2007. IEEE Int 2007:3329–3332Google Scholar
- Kang Y, Pan L, Sun M, Chen Q, Wang Y et al (2017) Gaussian mixture model based cloud detection for Chinese high resolution satellite imagery. Geomat Inf Sci Wuhan Univ 42(6):782–788Google Scholar
- Lin WT, Lin CH, Wu TH et al (2010) Image segmentation using the K-means algorithm for texture features. World Acad Sci Eng Technol 38(5):434–435Google Scholar
- Mohajerani S, Krammer T A, Saeedi P (2018) Cloud detection algorithm for remote sensing images using fully convolutional neural networks. https://doi.org/10.1109/MMSP.2018.8547095
- Ozkan S, Efendioglu M, Demirpolat C (2018) Cloud detection from RGB color remote sensing images with deep pyramid network. arXiv:1801.08706v1Google Scholar
- Shi M, Xie F, Zi Y, et al (2016) Cloud detection of remote sensing images by deep learning. IGARSS 2016–2016 IEEE international geoscience and remote sensing symposium. IEEEGoogle Scholar
- Wang H, Zhang R, Kuang J et al (2016) A progressive refinement scheme for cloud detection of aerial images. J Wuhan Univ (Nat Sci Edn) 62(6):525–530Google Scholar
- Xie F, Shi M, Shi Z et al (2017) Multilevel Cloud Detection in Remote Sensing Images Based on Deep Learning. IEEE J Sel Top Appl 2017:1–10Google Scholar
- Zhu H (2005) Digital image processing foundation. Science Press, BeijingGoogle Scholar