Share Email Print
cover

Proceedings Paper

Cloud concentration classification of UAV images based on image quality
Author(s): Song Xue; Si-Yu Zhang; Cong-Li Li
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

The UAV is easy to be affected by cloud when it is shooting to the ground. It shielding the ground infor-mation and reducing the image quality. It affecting the extraction of prior information and image pro-cessing. At present, there is no feasible and effective method for cloud concentration of cloud images. Therefore, This paper proposed a cloud concentration classification method based on the quality of cloud images. Based on the analysis of the structure of the image, 6 kinds of feature factors which are sensitive to the quality of the cloud images are extracted, and the feature vectors are constructed. And get the quality assessment model to obtain the quality score. Finally, the mean dif-ference of the Mahalanobis distance between the original image set and the test image is used to obtain the cloud images concentration. In view of the quality assessment model and the cloud concentration classification standard, the real-time test cloud images are used as the test database. The algorithm is veri-fied by three aspects: the subjective and objective consistency of image quality, the accuracy of cloud concentration classification, and the efficiency of algorithm. The experimental results show that the algo-rithm has higher accuracy, better subjective and objective consistency, and the classification of image cloud concentration level is more clear, and the algorithm runs more efficiently. The algorithm can meet the cloud UAV images quality assessment and cloud concentration classification.

Paper Details

Date Published: 9 August 2018
PDF: 8 pages
Proc. SPIE 10806, Tenth International Conference on Digital Image Processing (ICDIP 2018), 108065R (9 August 2018); doi: 10.1117/12.2502816
Show Author Affiliations
Song Xue, Army Academy of Artillery and Air Defense (China)
Si-Yu Zhang, Army Academy of Artillery and Air Defense (China)
Cong-Li Li, Army Academy of Artillery and Air Defense (China)


Published in SPIE Proceedings Vol. 10806:
Tenth International Conference on Digital Image Processing (ICDIP 2018)
Xudong Jiang; Jenq-Neng Hwang, Editor(s)

© SPIE. Terms of Use
Back to Top